JP5563044B2 - Method and apparatus for analyte measurement using large-scale FET arrays - Google Patents

Description

  The present disclosure relates to an inventive method and apparatus for detecting and measuring one or more analytes via an electronic sensor.

  Electronic devices and electronic components have found numerous applications in chemistry and biology (more generally life sciences), particularly in the detection and measurement of various aspects of chemical reactions and material composition. It was. One such electronic device is called an ion sensitive field effect transistor and is often referred to in the related literature as ISFET (or pHFET). ISFETs have been traditionally studied mainly by academic and research organizations to facilitate the measurement of the hydrogen ion concentration of a solution (generally expressed in pH).

  More specifically, an ISFET is an impedance conversion device that operates in a manner similar to a MOSFET (Metal Oxide Semiconductor Field Effect Transistor), and is specifically configured to selectively measure ion activity in solution (eg, , Hydrogen ions in solution are “analytes”). The theory of detailed mechanism of action of ISFET is “Thirty years of ISFETOLOGY: what happened in the past 30 years and what may happen in the next 30 years”, P. Bergveld, Sens. Actuators, 88 (2003), pp. 1 -20 ”, the publication of which is incorporated herein by reference.

FIG. 1 shows a cross-sectional view of a p-type ISFET 50 manufactured using a conventional CMOS (complementary metal oxide semiconductor) process. The manufacture of the p-type ISFET is based on the p-type silicon substrate 52, and an n-type well 54 that forms the body of the transistor is formed. Highly doped p-type (p ⁺ ) regions S and D constituting the source 56 and drain 58 of the ISFET are formed in the n-type well. A highly doped n-type (n ⁺ ) region B is also formed in the n-type well to provide a conductive body (bulk) connection 62 with the n-type well. An oxide layer 65 is applied over the source, drain and body connection regions, and openings are created through the oxide layer 65 to provide electrical connection (via electrical conductors) to these regions. . For example, metal contact 66 acts as a conductor that provides an electrical connection to drain 58, and metal contact 68 provides a common connection to source 56 and n-type well 54 through highly conductive body connection 62. Acts as a conductor. Polysilicon gate 64 is formed on the oxide layer located over region 60 of n-type well 54 between source 56 and drain 58. Oxide layer 65 is often referred to as “gate oxide” because it is disposed between polysilicon gate 64 and the transistor body (eg, n-type well).

Like a MOSFET, the operation of an ISFET is the charge resulting from the MOS (metal oxide semiconductor) capacitance formed by the polysilicon gate 64 between the source and drain, the gate oxide 65 and the n-type well region 60. Based on distribution modulation. When a negative voltage is applied to the gate and source regions (V _GS <0 volts), a “p-channel” 63 is created by depletion of electrons at the interface between region 60 and gate oxide 65. . This p-channel 63 extends between the source and drain, and current is transmitted through the p-channel when the gate-source potential V _GS is a negative voltage sufficient to attract holes from the source to the drain. . The gate-source potential at which channel 63 begins to conduct current is referred to as the transistor threshold voltage V _TH (the transistor conducts when V _GS has an absolute value greater than the threshold voltage V _TH ). The source is so named because it is the source of charge carriers flowing through channel 63 (holes for the p-channel), and similarly the drain is where the charge carriers leave channel 63.

In the ISFET 50 of FIG. 1, the n-type well 54 (transistor body) forces the same potential as the source 56 through the body connection 62 as seen in the metal contact 68 that connects to both the source 56 and the body connection 62. This biased connection (eg, V _SB = 0 volts) prevents forward biasing of the P ⁺ source region and n-type well, and thereby charge carriers to the location of region 60 where channel 63 can be formed. Making it easier to limit Some potential difference between the source 56 and the body / n-type well 54 (a non-zero source-body voltage V _SB ) affects the threshold voltage V _{TH of the} ISFET in a non-linear relationship and is also commonly referred to as the body effect. And unfavorable for many applications.

  As also shown in FIG. 1, the polysilicon gate 64 of the ISFET 50 is disposed in one or more additional oxide layers 75 disposed on the gate oxide 65 to form a “floating gate” structure 70. Joined with multiple metal layers. The floating gate structure is so named because it is electrically independent from other conductors associated with the ISFET, ie, sandwiched between the gate oxide 65 and the passivation layer 72. In the ISFET 50, the passivation layer 72 constitutes an ion sensitive film that increases the ion sensitivity of the device. That is, the presence of ions in the “analysis solution” 74 (solution containing the ions of interest) in contact with the passivation layer 72, particularly the detection region 78 on the floating gate structure 70, is between the source 56 and the drain 58. The electrical characteristics of the ISFET are changed to change the current flowing through the p-channel 63. The passivation layer 72 may include any one of different materials to facilitate detection for specific ions. For example, a passivation layer comprising silicon nitride or silicon oxynitride generally detects the hydrogen ion concentration in the analytical solution 74, while a passivation layer comprising polyvinyl chloride containing valinomycin detects potassium from the analytical solution. (Substances that are suitable for passivation layers and that detect other ions such as sodium, silver, iron, bromine, iodine, calcium and nitric acid are known).

With respect to ion sensitivity, the potential difference is commonly referred to as “surface potential” and increases at the solid / liquid interface between the passivation layer 72 and the detection region 78 according to the ion concentration in the detection region 78 resulting from a chemical reaction (eg, Usually accompanied by dissociation of the oxide surface groups by ions in the analysis solution proximate to the detection region 78). This surface potential affects the threshold voltage V _TH of the ISFET, and thus the threshold voltage V _{TH of the} ISFET varies with changes in the ion concentration in the analysis solution 74 proximate to the detection region 78.

FIG. 2 illustrates an electrical circuit diagram of the p-channel ISFET 50 shown in FIG. Referring again to FIG. 1, the reference electrode 76 (conventional Ag / AgCl electrode) in the analytical solution 74 determines the potential of the bulk of the analytical solution itself, and as shown in FIG. It is similar. In the ISFET linear or non-saturated operating region, the drain current _ID is given by:

_Where V _DS is the voltage between the drain and source, and β is a transconductance parameter (unit Amps / Volts ² ) given by:

In the above equation, μ represents carrier mobility, C _OX is the gate oxide capacitance per unit area, and the W / L ratio is the ratio of the width and length of the channel 63. When the reference electrode 76 provides an electrical reference or ground (V _G = 0 volt) and the drain current _ID and the drain-source voltage V _DS are constant, the change in the ISFET source voltage V _S is expressed by the equation (1) And directly follows the change of the threshold voltage _VTH . This can also be seen by rearranging equation (1) into the following equation:

Since the threshold voltage V _{TH of the} ISFET is sensitive to ion concentration as described above, the source voltage V _S according to equation (3) provides a signal that is directly related to the ion concentration in the analysis solution 74 proximate to the detection region 78 of the ISFET. . In a typical conventional ISFET for pH detection using a silicon nitride and silicon oxynitride passivation layer, a threshold voltage sensitivity ΔV _TH (eg, a threshold voltage associated with a change in pH of the analysis solution) of about 30 mV / pH to about 50 mV / pH. Change) is observed (the 298 Kelvin degree and the theoretical maximum threshold is 59.2 mV / pH).

Previous research efforts to produce ISFETs for pH measurement based on conventional CMOS process technology have aimed to obtain a linearly high signal over the pH 1-14 range. When using a typical threshold sensitivity of about 50 mV / pH, considering equation (3) above, this requires a linear operating range of about 700 mV as the source voltage V _S. As described above with reference to FIG. 1, the ISFET threshold voltage V _TH (as well as the MOSFET) is affected by any voltage V _SB between the source and the body (n-type well). More specifically, the threshold voltage _VTH is a non-linear function of a non-zero source-body (n-type well 54) voltage _VSB . Therefore, in order to prevent the loss of linearity due to the source-drain voltage potential difference (ie, to mitigate the “body effect”), the source connection 56 and the body connection 62 of the ISFET 50 are often metal as shown in FIG. It is connected to a common potential via a contact 68. This body-source connection is also shown in the electrical schematic of ISFET 50 shown in FIG.

  Prior art efforts to fabricate a two-dimensional array of ISFETs based on the ISFET design of FIG. 1 result in up to 256 ISFET sensor elements, or “pixels” (eg, a 16 pixel × 16 pixel array) in one array. It was. A typical study in ISFET array fabrication is the publication “A large transistor-based sensor array chip for direct extracellular imaging,” MJ Milgrew, MO Riehle, and DRS Cumming, Sensors and Actuators, B: Chemical, 111-112, ( 2005), pp. 347-353, and “The development of scalable sensor arrays using standard CMOS technology,” MJ Milgrew, PA Hammond, and DRS Cumming, Sensors and Actuators, B: Chemical, 103, (2004), pp. 37 This publication is incorporated herein by reference and is collectively referred to below as “Milgrew et al”. Other research efforts on the realization of ISFET arrays are published in the publication “A very large integrated pH-ISFET sensor array chip compatible with standard CMOS processes,” TCW Yeow, MR Haskard, DE Mulcahy, HI Seo and DH Kwon, Sensors and Actuators B : Chemical, 44, (1997), pp. 434-440, and “Fabrication of a two-dimensional pH image sensor using a charge transfer technique,” Hizawa, T., Sawada, K., Takao, H., Ishida, M., Sensors and Actuators, B: Chemical 117 (2), 2006, pp. 509-515, which publication is also incorporated herein by reference.

FIG. 3 illustrates a column 85 _j of a two-dimensional ISFET array designed by Milgrew et al. Column 85 _j includes 16 pixels from 80 ₁ to 80 ₁₆ , and, as will be described below with respect to FIG. 7, a complete two-dimensional array consists of 16 such columns 85 _j ( j = 1, 2, 3,... 16). As shown in FIG. 3, a given column 85 _j includes a current source I _SOURCEj shared by all pixels in the column and also an ISFET bias / read circuit 82 j (including current sink I _Sinkj ) shared by all pixels in the column. . Each pixel from 80 ₁ to 80 ₁₆ has a p-channel ISFET 50 (as shown in FIGS. 1 and 2) having an electrically connected source and drain, and an additional 16 rows of selection signals (RSEL ₁ to RSEL ₁₆ , and It includes two switches S1 and S2 that respond to one of its complements. As described below with respect to FIG. 7, the row selection signal and its complementary signal are generated simultaneously to “validate” or select a given pixel in column 85 _j , such a pair of signals consecutively different pixels in the column. In order to make them valid one by one, they are generated with an array.

As shown in FIG. 3, the switch S2 of each pixel 80 in the Milgrew et al design is implemented as a conventional n-channel MOSFET that connects the current source I _SOURCEj and the source of the ISFET ₅₀ upon receipt of the corresponding row selection signal. . The switch S1 of each pixel 80 is a CMOS that includes a transmission gate, ie, an n-channel MOSFET and a p-channel MOSFET that connect the source of the ISFET 50 and the bias / read circuit 82 _j when receiving the corresponding row select signal and its complementary signal. Implemented as a pair. FIG. 4 illustrates switch S1 ₁ of pixel 80 ₁ where the p-channel MOSFET of the transmission gate is shown as S1 _1P and the n-channel MOSFET is shown as S1 _1N . In the design of Milgrew et al, transmission gate is used for the switch S1 of the pixel, for a valid pixel, applied to the bias / readout circuitry 82 _j ISFET source voltage within the power supply range of _V DD _{~V SS} And can be output by the column as signal V _Sj . From the above, it should be understood that in the ISFET sensor array design of the Milgrew et al design, each pixel 80 includes four transistors, i.e., they are p-channel ISFET, n-channel for switch S1. CMOS pair transfer gate including MOSFET and p-channel MOSFET and n-channel MOSFET for switch S2.

Also, as shown in FIG. 3, the bias / read circuit 82 _j maintains a constant drain-source voltage V _DSj for the effective pixel ISFET in column 85 _j and from a constant drain current I _SOURCEj. In order to isolate the measurement of the source voltage V _Sj , a Kelvin bridge type source-drain follower configuration is used. To this end, the bias / readout circuitry 82 _j includes two operational amplifiers A1 and A2, having a current sink _{I Sinkj} and resistor _{R SDj.} Between the drain and the source of ISFET effective pixel as the source voltage V _DSj - voltage generated in both poles of the resistor R _SDj by a current sink I _Sinkj flowing through the resistor, the resistor operational amplifier, a constant drain Forced to occur. Thus, referring _again to equation (3), with constant V _DSj and constant I _SOURCEj , the effective pixel ISFET source voltage V _Sj provides a signal corresponding to the ISFET threshold voltage V _TH , and therefore ISFET detection. Provide a pH measurement in close proximity to the area (see FIG. 1). The wide dynamic range for the source voltage V _Sj obtained by the transmission gate S1 can reliably measure the entire range of pH 1-14, and the source-body connection of each ISFET is an ISFET over the entire pH measurement range. Ensure sufficient linearity of the threshold voltage.

In the Milgrew et al column design shown in FIG. 3, the p-channel ISFET 50 shown in FIG. 1 must be used for each pixel in order for the Kelvin bridge configuration of the column bias / read circuit 82 _j to function properly. Should be understood. More specifically, with n-channel ISFETs, alternative implementations based on Kelvin bridge configurations are not possible. Referring again to FIG. 1, for an n-channel ISFET based on a conventional CMOS process, the n-type well 54 is not required, and the highly doped n-type regions for the drain and source (form the transistor body) It is directly formed on the p-type silicon substrate 52. For n-channel ISFET devices, the transistor body is typically connected to electrical ground. In the design of Milgrew et al, given the requirement that the source and drain of the ISFET be electrically connected together to mitigate the nonlinear behavior due to body effects, this is because the source of the n-channel ISFET is also electrically grounded (eg, , V _S = V _B = 0 volts), which would prevent useful information from valid pixels. Thus, the Milgrew et al column design shown in FIG. 3 requires a p-channel ISFET for proper operation.

In the Milgrew et al column design shown in FIG. 3, the two n-channel MOSFETs required to implement switches S1 and S2 in each pixel are not formed in the n-type well 54 shown in FIG. No, where a p-channel ISFET is formed for the pixel. Rather, it should be understood that the n-channel MOSFET is formed directly on the p-type silicon substrate 52, outside the boundary of the n-type well 54 of the ISFET. FIG. 5 is a view similar to FIG. 1 and illustrates a portion of the p-type silicon substrate 52 corresponding to one pixel 80 in the column 85j shown in FIG. , The source 56 and the n-type well 54 including the body connection 62 of the ISFET 50 constitute the first n-channel MOSFET corresponding to the switch S2 and _one of the two transistors of the transmission gate S11 shown in FIG. Two n-channel MOSFETs _1N are shown side by side.

Further, in the design of Milgrew et al, a p-channel MOSFET that is required to implement a transmission gate S1 for each pixel (see, eg, S1 _1p in FIG. 4) forms a p-channel ISFET 50 for that pixel. Cannot be formed in the same n-type well. In particular, since the body and source of the p-channel ISFET are electrically connected, mounting the _p -channel MOSFET 1 _1p in the same n-type well as the p-channel ISFET 50 will cause unpredictable operation of the transmission gate. Or make all operations completely impossible. Thus, the Milgrew et al design requires that each pixel be implemented with two separate n-type wells. FIG. 6 is a view similar to FIG. 5 and illustrates a cross-sectional view of another portion of the p-type silicon substrate 52 corresponding to the pixel 80, where the n-type well 54 corresponding to the ISFET 50 is shown in FIG. Are shown side by side in the second n-type well 55 forming the p-channel MOSFET 1 _1P constituting one of the two transistors of the transmission gate S1 ₁ shown in FIG. The drawings in FIGS. 5 and 6 are not intended to be dimensional and do not strictly represent the particular pixel arrangement in the Milgrew et al design. Rather, these figures are essential conceptual diagrams and illustrate the requirements of a large number of n-type wells and n-channel MOSFETs fabricated outside the n-type well in the design of Milgrew et al. It is an object.

  Milgrew et al's array design was implemented using a conventional CMOS fabrication process of 0.35 micrometers (μm). In this process, various design rules determine the minimum inter-component distance. For example, according to the 0.35 micrometer CMOS design rule and referring to FIG. 6, the adjacent n-well distance “a” must be at least 3 micrometers. In FIG. 6, the distance “a / 2” is also shown on the left side of n-well 54 and n− to illustrate the minimum distance required to separate pixel 80 from the left and right adjacent pixels in the other column. Shown on the right side of well 55. Further, according to the 0.35 micrometer CMOS design rule, the distance “b” indicating the width in the cross-sectional view of the n-type well 54 and the distance “c” indicating the width in the cross-sectional view of the n-type well 55 shown in FIG. Are in the order of about 3 μm to 4 μm (within the n-type well, a clearance of 1.2 μm is created between the end of the n-well and the source and drain respectively, and the source and drain are in the order of 0.7 μm Therefore, the total length “d” shown in FIG. 6 showing the width of the pixel 80 in the cross-sectional view is on the order of about 12 μm to 14 μm. In one implementation, Milgrew et al discloses an array based on the column / base design shown in FIG. 3 that includes 12.8 μm × 12.8 μm geometric square pixels each.

  That is, the Milgrew et al ISFET pixel design aims to ensure the accuracy of hydrogen ion concentration measurements in the pH range 1-14. In order to ensure the linearity of the measurement, the source and body of each pixel are electrically connected. In order to ensure the full range of pH measurements, a transmission gate S1 is used in each pixel to convey the effective pixel source voltage. Thus, each pixel in the Milgrew array requires four transistors (p-channel ISFET, p-channel MOSFET and two n-channel MOSFETs) and two separate n-wells (FIG. 6). Based on the conventional 0.35 micrometer CMOS manufacturing process and corresponding design rules, the pixels of such an array will have a minimum size well above 10 μm, ie on the order of about 12 μm to 14 μm.

FIG. 7 shows the entire two-dimensional pixel array 95 according to the design of Milgrew et al with row and column decoder circuitry and measurement readout circuitry. The array 95 includes 16 columns of pixels 85 ₁ to 85 ₁₆ , and each column includes 16 pixels as described above with respect to FIG. 13 (eg, a 16 pixel × 16 pixel array). Row decoder 92 provides a complementary pair of row select signals, each pair of row select signals being a column output from array 95 based on the respective source voltage V _S1 -V _S16 of the valid row of ISFETs. To provide a signal, one pixel in each of the columns 85 _{1 to} 85 ₁₆ is enabled simultaneously. Row decoder 92 is implemented as a conventional 4-16 decoder (e.g., 4-bit binary input _ROW 1 -row ₄ for selecting one of the ^{2 4} outputs). The set of column output signals V _S1 -V _S16 for the valid rows of the array is applied to switching logic 96 that includes ₁₆ transmission gates S ₁ -S ₁₆ (one transmission gate for each output signal). As described above, the transmission gate of the switching logic 96 is implemented using a p-channel MOSFET and an n-channel MOSFET to ensure a sufficient dynamic range for each output signal V _{S1 to} V _S16 . The column decoder 94, similar to the row decoder 92, is implemented as a conventional 4-16 decoder and provides the switching logic 96 at any given time to provide a single output signal V _S from the switching logic 96. Controlled via a 4-bit binary input COL ₁ -COL ₄ that enables one of the transmission gates S ₁ -S ₁₆ . The output signal V _S is applied to a 10-bit analog to digital converter (ADC) 98 to provide a digital representation D ₁ -D ₁₀ of the output signal Vs corresponding to a given pixel of the array.

  As noted above, individual ISFETs and ISFET arrays are similar to those described above and are used as sensing devices in a variety of chemical and biological applications. In particular, ISFETs are used as pH sensors in various processes involving nucleic acids such as DNA. Several examples of using ISFETs in various life science applications are described in the following publications, each of which is incorporated herein by reference: Massimo Barbaro, Annalisa Bonfiglio, Luigi Raffo, Andrea Alessandrini , Paolo Facci and Imrich Barak, “Fully electronic DNA hybridization detection by a standard CMOS biochip,” Sensors and Actuators B: Chemical, Volume 118, Issues 1-2, 2006, pp. 41-46, Toshinari Sakurai and Yuzuru Husimi, “ Real-time monitoring of DNA polymerase reactions by a micro ISFET pH sensor, ”Anal. Chem., 64 (17), 1992, pp 1996-1997, S. Purushothaman, C. Toumazou, J. Georgiou,“ Towards fast solid state DNA sequencing, ”Circuits and Systems, vol.4, 2002, pp. IV-169 to IV-172, S. Purushothaman, C. Toumazou, CP Ou,“ Protons and single nucleotide polymorphism detection: A simple use for the Ion Sensitive Field Effect Transistor, ”Sensors and Actuators B: Chemical, Vol. 114, no 2, 2006, pp. 964-968, AL Simonian, AW Flounders, JR Wild, “FET-Based Biosensors for The Direct Detection of Organophosphate Neurotoxins,” Electroanalysis, Vol. 16, No. 22, 2004, pp. 1896- 1906, C. Toumazou, S. Purushothaman, “Sensing Apparatus and Method,” United States Patent Application 2004-0134798, published July 15, 2004, and TW Koo, S. Chan, X. Su, Z. Jingwu, M. Yamakawa , VM Dubin, “Sensor Arrays and Nucleic Acid Sequencing Applications,” United States Patent Application 2006-0199193, published September 7, 2006.

  In general, the development of rapid and sensitive nucleic acid sequencing methods utilizing DNA sequences has greatly advanced our understanding of biology. The term “sequencing” refers to determining the primary structure (or primary sequence) of an unbranched biopolymer, which briefly summarizes most of the atomic structure of the sequenced molecule. Gives a linear representation of the symbol, known as “DNA sequencing” refers specifically to the process of determining the nucleotide sequence of a given DNA fragment. Currently, analysis of the entire genome of viruses, bacteria, fungi, animals and plants is possible, but such analysis is generally limited by cost and sequencing throughput. More specifically, current conventional sequencing methods are limited in terms of sequence accuracy, length of individual templates that can be sequenced, sequence cost, and sequencing speed.

  Despite advances in sample preparation and sequencing technology, current conventional sequencing strategies, including those that include ISFETs, deliver throughput to the level required for analysis of large numbers of individual human genomes. It does not provide the cost savings needed to raise it. In order to understand the genetic basis of disease and aging, it is necessary to sequence a large number of individual genomes. A number of cancers will also need to be sequenced to understand the changes in somatic cells that cause cancer. Several recent efforts have greatly improved the ability to prepare genomes for sequencing and to simultaneously sequence multiple templates. However, these or other approaches are intended to read the relatively large size reaction volumes needed to prepare templates that can be detected by these systems, as well as the need for specialized nucleotide analogs and bases. Limited by complex enzymes or fluorescent methods.

  The inventors have recognized and understood that ISFET large arrays can be specially constructed and used to promote DNA sequencing technology based on monitoring changes in chemical processes related to DNA synthesis. . More generally, the inventors have found that large arrays of chemosensitive FETs can be used in a variety of hosts in chemical and / or biological processes (chemical reactions, cell culture, neural activity, nucleic acid sequencing, etc.). Recognize and understand that it can be used to detect analytes (eg, hydrogen ions, other ions, non-ionic molecules or compounds, binding events, etc.) and measure concentration / level, where such analysis Useful information can be obtained based on the measurement of objects.

  Accordingly, various aspects herein relate generally to inventive methods and apparatus for large-scale FET arrays for measuring one or more analytes. In various aspects disclosed herein, the FET array includes a number of “chemFET” or chemically sensitive field effect transistors that act as chemical sensors. The ISFET described above is one characteristic type of chemFET configured for ion detection, and the ISFET may be used in various ways as disclosed herein. Other types of chemFETs contemplated by this specification include ENFETs, which are configured for the detection of special enzymes. However, it should be understood that the present disclosure is not limited to ISFETs and ENFETs, and more generally relates to any FET configured for some type of chemical sensitivity.

  According to yet another aspect, the present disclosure generally relates to inventive methods and apparatus relating to transporting a suitable chemical sample to cause a corresponding reaction to the large-scale chemFET array. To facilitate the determination of chemical (eg, ions or other components) concentration or other measurements in an analyte at high speed, high density, a chemical sample contains a small amount of reaction (liquid) analyte be able to.

  For example, some aspects relate to “very large” two-dimensional chemFET sensor arrays (eg, beyond 256k sensors), where one or more independent chemical reactions or events that occur in close proximity to the array pixels are monitored. To do so, a chemFET is constructed that includes the elements or “pixels” that make up the sensor of such an array. In some exemplary implementations, the array is analyzed between one or more reaction chambers or microfluidic structures that form “wells” or “microwells” on individual sensors or groups of sensors, and measurements. The sample may be connected to a device for transporting to and removing from the well. Even when microwells are not used, the sensor array may be connected to one or more microfluidic structures for transporting analytes to pixels or removing analytes between measurements. Accordingly, the inventive aspects of the present disclosure, the aspects to be protected, are various microfluidic structures that can be used to flow reagents / analytes toward and away from wells or pixels, wells, methods of making arrays, Methods and structures for connecting wells to array pixels and methods and apparatus for loading DNA-bearing beads into wells when the apparatus is used for DNA sequencing or related analysis are included.

  A unique reference electrode and its connection with the flow cell are also shown.

  In various embodiments, the analyte of particular interest is hydrogen ions, and large ISFET arrays according to the present disclosure are specifically configured for pH measurements. In other embodiments, the monitored chemical reaction may be related to a DNA synthesis process, or other chemical and / or biological process, and the chemFET array is specifically pH or certain interesting. It may be configured to measure one or more analytes that provide relevant information about the chemical reaction. In various aspects, the chemFET array is manufactured using conventional CMOS fabrication technology and is specifically configured to facilitate rapid data collection from the entire array (to obtain the corresponding pixel output signal, all Scan pixels).

  With respect to analyte detection and measurement, in various aspects described in the more detailed description below, one or more analytes measured by a chemFET array according to the present disclosure may be a chemical reaction or a chemical process of interest (eg, multiple processes). Any of a variety of chemicals that provide relevant information regarding the binding of nucleic acid strands, binding of antibodies to antigens, etc. In some aspects, the ability to measure the level or concentration of one or more chemicals provides valuable information related to chemical or other processes in addition to simply detecting the presence of an analyte To do. In other aspects, mere detection of the presence of one analyte or the analyte of interest can provide valuable information.

  ChemFETs according to various inventive aspects of the present disclosure may be configured with high sensitivity to one or more diverse analytes / chemicals. In one embodiment, one or more chemFETs of the array may be configured for sensitivity to one or more analytes that exhibit one or more binding events (eg, for nucleic acid sequencing processes). In other embodiments, a given array of chemFETs may be configured for sensitivity to different analytes. For example, in one embodiment, one or more sensors (pixels) of the array may include a first type of chemFET configured to be chemically sensitive to the first analyte, and One or more other sensors in the array may include a second type of chemFET configured to be chemically sensitive to a second analyte that is different from the first analyte. Of course, it should be understood that more than two different chemFETs may be used in any given array for detecting or measuring different types of analyte / binding events. In general, in any of the embodiments of sensor arrays presented herein, a given array is “homogeneous” to detect the same type of analyte (eg, pH or other ion concentration) and / or It may include substantially similar or identical chemFETs for measurement, or the sensor array may be “heterogeneous” and include heterogeneous chemFETs to detect and / or measure dissimilar analytes. Should be understood.

  In yet another aspect, the inventors have significantly reduced the pixel size, thereby increasing the number of chemFET array pixels for a given semiconductor die size (ie, increasing the pixel density). Special improvements were made to the Milgrew et al ISFET array design described above with respect to 1-7, as well as other conventional ISFET arrays. In various aspects, this increase in pixel density is accompanied by an increase in the signal-to-noise ratio (SNR) of the output signal corresponding to the respective measurement results for the monitored chemical process. In particular, we have relaxed the chemFET linearity requirements and narrowed it to a more limited measured output signal range (eg, an output corresponding to a pH range of about 7-9 rather than 1-14). Signal), the complexity and size of individual pixels is greatly reduced, which is recognized and understood to facilitate the realization of ultra-large scale high density chemFET arrays. We also have alternative approaches to pixel selection in chemFET arrays (eg, the row and column decoder approach used in the Milgrew et al design shown in FIG. 7 that increases in complexity with array size). Recognizes and understands that it facilitates rapid data collection from very large, dense arrays.

With respect to chemFET array fabrication, the inventors have further described several techniques used in conventional CMOS fabrication processes as well as various post-fabrication process steps (wafer handling, cleaning, dicing, packaging, etc.). Recognizes and understands the negative impact on the resulting chemFET array. For example, referring again to FIG. 1, one potential problem is that trapped charges that may occur in the gate oxide 65 during etching of the metal associated with the floating gate structure 70, and such trapped charges may be reduced to the chemFET threshold. It relates to how the voltage V _TH can be affected. Another potential problem relates to the density / gap of the chemFET passivation layer (see, eg, passivation layer 72 of FIG. 1) resulting from the low temperature material deposition process commonly used in aluminum metal based CMOS fabrication. Such low temperature processes generally provide suitable passivation layers for conventional CMOS devices, but they are low density and porous passivation that can potentially be a problem for chemFETs when contacted by an analytical solution. Can result in layers. In particular, low-density porous passivation layer begins to absorb the analyte or other substances in the solution with time saturation can cause undesirable time-dependent drift in chemFET threshold voltage V _TH. This phenomenon hinders accurate measurement of one or more analytes of interest. In view of the foregoing, other inventive aspects disclosed herein provide a method and apparatus for mitigating potential adverse effects on chemFET functionality that may arise from various aspects of chemFET array fabrication and post-manufacturing processes / handling. About.

  Accordingly, one aspect of the present invention includes an array of CMOS-fabricated sensors, each sensor including one chemically-sensitive field effect transistor (chemFET), 10 μm at the array surface. The present invention relates to an apparatus occupying an area of 10 μm or less.

  Another aspect includes a two-dimensional array comprising at least 512 rows and at least 512 columns of electronic sensors, each sensor providing an output signal indicative of the presence and / or concentration of an analyte proximate to the two-dimensional array surface. The present invention relates to a sensor array including a configured chemically sensitive field effect transistor (chemFET).

  Another aspect relates to an apparatus that includes an array of CMOS sensors, each sensor including one chemically sensitive field effect transistor (chemFET). An array of CMOS sensors includes more than 256 sensors, and a collection of chemFET output signals from all chemFETs in the array constitutes a data frame. The apparatus further includes a control circuit configured to generate at least one array output signal and connected to the array to provide multiple data frames from the array at a frame rate of at least 1 frame / second. In one aspect, the frame rate may be at least 10 frames / second. In another aspect, the frame rate may be at least 20 frames / second. In yet other aspects, the frame rate may be at least 30, 40, 50, 70 or up to 100 frames / second.

  Another aspect relates to an apparatus that includes an array of CMOS sensors, each sensor including one chemically sensitive field effect transistor (chemFET). The chemFET includes a floating gate structure and a source and a drain having a first semiconductor type provided in a region having a second semiconductor type. Here, the region having the second semiconductor type and the source or drain are arranged. There is no electrical conductor to connect electrically.

  Another aspect relates to an apparatus comprising three field effect transistors (FETs), including an electronic sensor array, each sensor including one chemosensitive field effect transistor (chemFET).

  Another aspect relates to an apparatus that includes an electronic sensor array, wherein each sensor includes no more than three field effect transistors (FETs), and no more than three FETs include one chemosensitive field effect transistor (chemFET).

Another aspect includes an electronic sensor array, wherein each sensor includes a plurality of field effect transistors (FETs) including one chemosensitive field effect transistor and a plurality of electrical conductors electrically connected to the plurality of FETs; Here, a plurality of FETs relates to a device in which a plurality of electrical conductors are arranged to include no more than four conductors that traverse the area occupied by each sensor and interconnect with a number of sensors in the array.

  Another aspect includes an array of CMOS sensors, each sensor including a plurality of field effect transistors (FETs) including one chemosensitive field effect transistor (chemFET), and all FETs of each sensor are of the same channel type. And relates to an apparatus mounted on a single semiconductor region of an array substrate.

  Another aspect relates to a sensor array that includes a plurality of electronic sensors arranged in a plurality of rows and a plurality of columns. Each sensor includes one chemosensitive field effect transistor (chemFET) and is configured to provide at least one output signal indicative of the presence and / or concentration of an analyte proximate to the array surface. Includes a transistor (chemFET). For each column of the plurality, the array includes a column circuit configured to provide a constant drain current and a constant drain-source voltage to each chemFET in the column, the column circuit comprising two operational amplifiers and To provide a constant drain-source voltage, each chemFET and a diode connected FET arranged in a Kelvin bridge configuration are included.

  Another aspect relates to a sensor array that includes a plurality of electronic sensors arranged in a plurality of rows and a plurality of columns. Each sensor includes one chemosensitive field effect transistor (chemFET) and is configured to provide at least one output signal indicative of the concentration of ions in the analyte proximate the array surface. (ChemFET). The array further includes at least one row selection shift register that enables each of the plurality of rows and at least one column shift register that obtains an output signal from each of the plurality of columns.

  Another aspect relates to an apparatus that includes an array of CMOS sensors, each sensor including one chemically sensitive field effect transistor (chemFET). The chemFET includes a floating gate structure and a source and drain having a first semiconductor type provided in a region having a second semiconductor type, where the region having the second semiconductor type and the source or drain are arranged. There is no electrical conductor to connect electrically. The array includes a two-dimensional array of CMOS sensors of at least 512 rows and at least 512 columns. Each sensor consists of three field effect transistors (FETs) including one chemically sensitive field effect transistor (chemFET), and each sensor includes a plurality of electrical conductors electrically connected to the three FETs. The three FETs are arranged such that the plurality of electrical conductors includes no more than four conductors that traverse the area occupied by each sensor and interconnect with multiple sensors in the array. All FETs of each sensor are of the same channel type and are mounted on a single semiconductor region of the array substrate. The collection of chemFET output signals from all chemFETs in the array builds a data frame. The apparatus further includes a control circuit connected to the array configured to construct at least one array output signal to provide multiple data frames from the array at a frame rate of at least 20 frames / second. .

  Another aspect relates to a method of manufacturing an array of CMOS sensors, each sensor including a chemically sensitive field effect transistor (chemFET). The method includes: A) dicing the semiconductor wafer to form at least one diced portion comprising an array, and B) forming gas annealing on the at least one diced portion.

  Another aspect relates to a method of manufacturing an array of CMOS sensors. Each sensor includes a chemically sensitive field effect transistor (chemFET) having a silicon nitride and / or silicon oxynitride chemically sensitive passivation layer deposited by plasma enhanced chemical vapor deposition (PECVD). The method includes: A) depositing at least one additional passivation material on the chemically sensitive passivation layer for decreasing porosity and / or increasing density of the passivation layer.

  In another aspect, the invention places a plurality of template nucleic acids in a plurality of reaction chambers, wherein the plurality of reaction chambers are at least one chemosensitive field effect transistor for each reaction chamber. A chemFET array comprising (chemFET) and each template nucleic acid hybridizes to a sequencing primer and binds to a polymerase) one or more known nucleotide triphosphates, Detect the incorporation of one or more known nucleotide triphosphates by synthesizing a new nucleic acid strand by sequentially incorporating it into the 3 'end of the sequencing primer and by changing the current in at least one chemFET in the array Nucleic acid sequencing method comprising To provide.

  In another aspect, the invention places a plurality of template nucleic acids in a plurality of reaction chambers, wherein the plurality of reaction chambers is at least one chemosensitive field effect transistor (chemFET) for each reaction chamber. And each template nucleic acid is hybridized to a sequencing primer and bound to a polymerase), in order to sequence one or more known nucleotide triphosphates. Synthesizing a new nucleic acid strand by incorporating it at the 3 ′ end of the sing primer, detecting the incorporation of one or more known nucleotide triphosphates by changing the current in at least one chemFET in the array, A method of nucleic acid sequencing comprising chemFET array to provide what is either of the array.

  In another aspect, the invention places a plurality of template nucleic acids in a plurality of reaction chambers, wherein the plurality of reaction chambers is at least one chemosensitive field effect transistor (chemFET) for each reaction chamber. And wherein each template nucleic acid hybridizes to a sequencing primer and binds to a polymerase), one or more known nucleotide triphosphates in turn Synthesize a new nucleic acid strand by incorporation into the 3 ′ end of the sequencing primer, and detect incorporation of one or more known nucleotide triphosphates by generation of a sequencing reaction byproduct. A method of nucleic acid sequencing comprising: chemFET array There do include sensors that exceeds (a) 256, or (b) the center-to-center distance between adjacent chambers (or pitch) to provide what is 1 to 10 [mu] m.

Various aspects apply equally to the methods disclosed herein, and they are described once for the sake of brevity. In some embodiments, the center-to-center distance between adjacent chambers is 2-9 μm, such as about 2 μm, about 5 μm, or about 9 μm. In some embodiments, chemFET array sensor more than 256 (and optionally more than 256 corresponding reaction chambers (or wells), the sensor more than 10 ³ (and corresponding reaction chambers optionally greater than 10 ^3), 10 more than ⁴ sensors (and corresponding reaction chambers optionally more than 10 ^4), sensors of more than 10 ⁵ (and corresponding reaction chambers optionally greater than 10 ^5), more than 10 ⁶ to the sensor (and optionally more than 10 ⁶ In some embodiments, the chemFET array includes at least 512 rows and at least 512 columns of sensors.

  In some embodiments, the sequencing reaction byproduct is inorganic pyrophosphate (PPi). In some embodiments, PPi is measured directly. In some embodiments, PPi is measured in the absence of PPi receptor. In some embodiments, the sequencing reaction byproduct is a hydrogen ion. In some embodiments, the sequencing reaction byproduct is inorganic phosphoric acid (Pi). In other aspects, as shown herein, the chemFET detects changes in any combination of by-products, optionally in combination with other parameters.

  In another aspect, the invention places a plurality of template nucleic acids in a plurality of reaction chambers, wherein the plurality of reaction chambers is at least one chemosensitive field effect transistor (chemFET) for each reaction chamber. And wherein each template nucleic acid hybridizes to a sequencing primer and binds to a polymerase), one or more known nucleotide triphosphates in turn Synthesizing a new nucleic acid strand by incorporation into the 3 ′ end of the sequencing primer, directly detecting the release of inorganic pyrophosphate as an indicator of incorporation of one or more known nucleotide triphosphates. Nucleic acid sequencing methods comprising are provided.

  In some embodiments, PPi is detected directly by binding to a PPi receptor immobilized on a chemFET. In some embodiments, PPi is detected directly by chemFET in the absence of PPi receptors.

  In another aspect, the invention provides for fragmenting a template nucleic acid to produce a plurality of fragmented nucleic acids, and for producing a plurality of beads each having a single-stranded fragmented nucleic acid attached thereto. In addition, a single strand from a fragmented nucleic acid is individually attached to a bead, and a plurality of beads to which a single-stranded fragmented nucleic acid is attached, each sensor in the area has a separate reaction chamber. A method for nucleic acid sequencing comprising providing a chemFET array and performing a sequencing reaction in multiple chambers simultaneously is provided.

  In another aspect, the present invention provides a device including a chemically sensitive field effect transistor (chemFET) having a PPi receptor disposed on a surface.

  In some embodiments, the PPi selective receptor is Compound 1, Compound 2, Compound 3, Compound 4, Compound 5, Compound 6, Compound 7, Compound 8, Compound 9 or Compound 10 shown in FIG. 11B. In some embodiments, the chemFET is present on a chemFET array in which PPi selective receptors are disposed on each surface. In some embodiments, the same PPi selective receptor is placed on each chemFET of the array. In some embodiments, the array includes more than 256 sensors. In some embodiments, the array includes at least 512 rows and at least 512 columns of sensors. In some embodiments, the chemFET array is located at the bottom of the reaction chamber.

  In another aspect, the present invention provides an apparatus comprising a chemically sensitive field effect transistor (chemFET) having a biological array disposed on a surface.

Biological arrays may be nucleic acid arrays, protein arrays including but not limited to enzyme arrays, antibody arrays and antibody fragment arrays, cell arrays, and the like. The chemical array may be, but is not limited to, an organic small molecule array or an inorganic molecule array. The chemFET array may include at least ⁵ , 10, 10 ² , 10 ³ , 10 ⁴ , 10 ⁵ , 10 ⁶ or more sensors. In some embodiments, the biological or chemical array is arranged in multiple “cells” or spatially defined regions, each region located on a different sensor in the chemFET array.

  In yet another aspect, the present invention contacts a nucleic acid array disposed on a chemFET array with a sample, and detects binding between the sample-derived nucleic acid and one or more regions on the nucleic acid array. A method for detecting a nucleic acid comprising:

  In another aspect, the invention contacts a protein array disposed on a chemFET array with a sample, and detects binding of the sample-derived protein to one or more regions on the protein array; A method for detecting a protein comprising

  In yet another aspect, the present invention contacts a protein array disposed on a chemFET array with a sample, and detects binding of nucleic acid from the sample to one or more regions on the protein array. A method for detecting a nucleic acid comprising:

  In another aspect, the invention contacts a sample with an antibody array disposed on a chemFET array, and detects binding of a sample-derived antigen to one or more regions on the antibody array; A method for detecting an antigen comprising is provided.

  In another aspect, the present invention contacts a sample with an enzyme array disposed on a chemFET array, and detects binding between a sample-derived target substance and one or more regions on the enzyme array. A method for detecting an enzyme substrate or inhibitor comprising:

  It is understood that all combinations of the above concepts and further concepts discussed in more detail below (assuming such concepts provided are not inconsistent with each other) are intended as part of the subject matter disclosed herein. It should be. In particular, all combinations of claimed subject matter are contemplated as part of the inventive subject matter disclosed herein. It is also to be understood that the terminology used explicitly herein, which appears in the disclosure incorporated by reference, is given the meaning most consistent with the special concepts disclosed herein.

  In the drawings, reference numerals used throughout the different drawings refer to the same parts. The drawings are not necessarily to scale, emphasis being placed on generally illustrating the various concepts presented herein.

FIG. 1 illustrates a cross-sectional view of a p-type ion-sensitive field effect transistor (ISFET) manufactured using a conventional CMOS process.

FIG. 2 is a diagram illustrating an electrical circuit diagram of the p-channel ISFET shown in FIG.

FIG. 3 is a diagram illustrating one column of the two-dimensional array based on the ISFET shown in FIG.

FIG. 4 is a diagram illustrating a transmission gate including a p-channel MOSFET and an n-channel MOSFET used for each pixel of the array column shown in FIG.

FIG. 5 is a view similar to FIG. 1, illustrating a portion of the substrate corresponding to one pixel of the array column shown in FIG. 3 in a larger cross-sectional view, where the ISFET is also included within the pixel. The two n-channel MOSFETs are shown side by side.

6 is a view similar to FIG. 5 and illustrating a cross-sectional view of a portion of the substrate corresponding to one pixel of the array column shown in FIG. 3, where ISFET is the transfer gate shown in FIG. Of p-channel MOSFETs.

FIG. 7 shows an overall two-dimensional pixel array based on the column design of FIG. 3 with row and column decoder circuitry and measurement readout circuitry.

FIG. 8 is a diagram generally illustrating a nucleic acid processing system that includes a large-scale chemFET array according to one inventive aspect of the present disclosure.

FIG. 9 is a diagram illustrating one column of chemFETs similar to the array shown in FIG. 8 according to one inventive aspect of the present specification.

FIG. 9A is a diagram illustrating a circuit diagram of an exemplary amplifier used in the array column shown in FIG. 9 according to one inventive aspect of the present disclosure, and FIG. 9B is a graph of amplifier bias and bandwidth. FIG. 9B is a graph of amplifier bias and bandwidth according to one inventive aspect of the present disclosure.

FIG. 10 is a diagram illustrating a plan view of a chip layout design for the pixels of the chemFET array column shown in FIG. 9 according to one inventive aspect of the present disclosure.

11A is a diagram illustrating a composite cross-sectional view along the line I-I of the pixel shown in FIG. 10, including additional elements between lines II-II and III-III in the right half of FIG. 1 illustrates an inter-layer diagram of pixel fabrication according to one inventive aspect of the present disclosure.

FIG. 11B-1 shows 10 PPi receptor compounds (compounds 1 to 4). FIG. 11B-2 shows 10 PPi receptor compounds (Compound 10). FIG. 11B-3 shows 10 PPi receptor compounds (compounds 5-9).

FIG. 11C-1 is a schematic diagram of the synthesis protocol for compound 4 of FIG. 11B-1. FIG. 11C-2 is a schematic diagram of the synthesis protocol for compound 4 of FIG. 11B-1. FIG. 11C-3 is a schematic diagram of the synthesis protocol for compound 4 of FIG. 11B-1.

FIG. 11D-1 is a diagram illustrating types of chemical substances that can be applied to the passivation layer to bind molecular recognition compounds (such as but not limited to PPi receptors). FIG. 11D-2 illustrates the types of chemicals that can be applied to the passivation layer to bind molecular recognition compounds (such as but not limited to PPi receptors). FIG. 11D-3 illustrates the types of chemicals that can be applied to the passivation layer to bind molecular recognition compounds (such as but not limited to PPi receptors).

FIG. 11E illustrates the adhesion of the compound 10 of FIG. 11B-2 to the metal oxide surface.

12A-12L are diagrams illustrating plan views of the fabrication layers shown in FIG. 11A according to one inventive aspect of the present disclosure.

FIG. 13 illustrates a block diagram of a typical CMOS IC chip implementation of a chemFET sensor array similar to FIG. 8, based on the column and pixel design shown in FIGS. 19-20 according to one inventive aspect of the present disclosure. It is.

FIG. 14 is a diagram illustrating the row select shift register shown in FIG. 13 according to one inventive aspect of the present disclosure.

15 is a diagram illustrating one of two of the column select shift registers of the array shown in FIG. 13 according to one inventive aspect of the present disclosure.

FIG. 16 is a diagram illustrating one of two of the output devices of the array shown in FIG. 13 according to one inventive aspect of the present disclosure.

FIG. 17 is a diagram illustrating a block diagram of the chemFET sensor array shown in FIG. 13 connected to an array controller, in accordance with one inventive aspect of the present disclosure.

18 is a diagram illustrating exemplary timing diagrams for various signals provided by the array controller of FIG. 17 according to one inventive aspect of the present disclosure.

FIG. 19 illustrates a block diagram of an alternative CMOS IC chip implementation of a chemFET sensor array according to another inventive aspect of the present disclosure. FIG. 20 is a diagram illustrating a block diagram of an alternative CMOS IC chip implementation of a chemFET sensor array according to another inventive aspect of the present disclosure.

20A is a diagram illustrating a top view of a chip layout design for the pixels of the chemFET array shown in FIG. 20 according to another inventive aspect of the present disclosure.

FIG. 21 is a diagram illustrating a block diagram of a further alternative CMOS IC chip implementation of a chemFET sensor array according to another inventive aspect of the present disclosure. FIG. 22 is a diagram illustrating a block diagram of a further alternative CMOS IC chip implementation of a chemFET sensor array according to another inventive aspect of the present disclosure. FIG. 23 is a block diagram illustrating a further alternative CMOS IC chip implementation of a chemFET sensor array according to another inventive aspect of the present disclosure.

24 is a diagram illustrating the pixel design of FIG. 9 with an n-channel chemFET implemented and an n-channel MOSFET according to another inventive aspect of the present disclosure.

FIG. 25 is a diagram illustrating an alternative pixel design and associated column column according to another inventive aspect of the present disclosure. FIG. 26 is a diagram illustrating an alternative pixel design and associated column column according to another inventive aspect of the present disclosure. FIG. 27 is a diagram illustrating an alternative pixel design and associated column column according to another inventive aspect of the present disclosure.

FIG. 28A is an isometric view of a portion of the microwell array used in the present disclosure to supplement the three-dimensional visualization of the array structure, showing circular and square wells. FIG. 28B is an isometric view of a portion of a microwell array used in the present disclosure to supplement the three-dimensional visualization of the array structure, showing circular and square wells.

FIG. 29 shows a top view of one corner (ie, the lower left corner) of the chip layout showing the individual ISFET sensor arrays on the CMOS die.

FIG. 30 is a diagram illustrating a layout of a mask (usually chromium) portion for one sensor / one well mode corresponding to a part of the die shown in FIG.

FIG. 31 is a diagram corresponding to the mask layout for the 4 sensor / well mode.

FIG. 32 illustrates the area surrounding the array to build a resist collar or wall (or basin (using the term in geological sense)) that surrounds the active array of on-substrate sensors shown in FIG. 33A. It is the figure which illustrated the 2nd mask used in order to mask.

FIG. 33 is an illustration of the generated basin.

FIG. 33A is an example of a three-layer PCM process for manufacturing a microwell array.

FIG. 34 schematically illustrates a suitable experimental apparatus incorporating a sensor array in the fluid interface. 35 is a cross-sectional view of the apparatus of FIG. 34 taken along line 35-35 '. FIG. 36 is an enlarged view of FIG. 35 by a perspective method. FIG. 37 is a further enlarged view of a part of the structure in order to make the fluid flow more visible.

FIG. 38 is a schematic diagram of a substrate with an etched photoresist that is the initial stage of formation in an example flow cell of a particular configuration.

FIG. 39 is a schematic diagram of a mask suitable for providing a first configuration of a flow cell consistent with FIG. FIG. 40 is a schematic view of a mask suitable for providing a first configuration of a flow cell consistent with FIG. FIG. 41 is a schematic diagram of a mask suitable for providing a first configuration of a flow cell consistent with FIG.

42 is a partial isometric cross-sectional view of the example device, showing the method and form of introducing a reference electrode, flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 43 is a partial isometric cross-sectional view of an example device showing the method and manner of introducing a reference electrode, the flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 44 is an enlarged view of FIG. 43 showing the method and form of introducing a reference electrode, the flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 45 is a partial isometric cross-sectional view of an example apparatus, showing the method and manner of introducing a reference electrode, flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 46 is an enlarged view of FIG. 45, showing the method and form of introducing the reference electrode, the flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 47 is a partial isometric cross-sectional view of an example device, showing the method and manner of introducing a reference electrode, flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 48 is an enlarged view of FIG. 47 showing the method and form of introducing a reference electrode, the flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 49 is a partial isometric cross-sectional view of an example device, showing the method and manner of introducing a reference electrode, flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 50 is an enlarged view of FIG. 49, showing the method and form of introducing a reference electrode, flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 51 is a partial isometric cross-sectional view of an example device, showing the method and manner of introducing a reference electrode, flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 52 is an enlarged view of FIG. 51 showing the method and form of introducing the reference electrode, the flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 53 is a partial isometric cross-sectional view of an example apparatus showing the method and manner of introducing a reference electrode, flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 54 is an enlarged view of FIG. 53, showing the method and form of introducing the reference electrode, the flow cell and flow chamber, and the use of materials such as plastic and PDMS.

FIG. 55 is a cross-sectional view of a two-layer glass (or plastic) arrangement for building a fluidic device that mounts on a chip for use as described herein. FIG. 56 is a cross-sectional view of a two-layer glass (or plastic) arrangement for constructing a fluidic device for mounting on a chip for use as described herein. FIG. 57 is a partial isometric cross-sectional view of an example device, showing the method and manner of introducing a reference electrode, flow cell and flow chamber, and the use of materials such as plastic and PDMS. FIG. 58 is an enlarged view of FIG. 57 showing the method and form of introducing the reference electrode, the flow cell and flow chamber, and the use of materials such as plastic and PDMS.

FIG. 59A is a diagram illustrating parts for two examples of two-piece injection molded parts to form a flow cell. FIG. 59B illustrates parts for two examples of two-piece injection molded parts to form a flow cell. FIG. 59C illustrates parts for two examples of two-piece injection molded parts to form a flow cell.

FIG. 60 is a schematic cross-sectional view for introducing a stainless steel capillary as an electrode downstream of a flow cell such as the flow cell of FIGS. 59A-59C or other flow cells.

FIG. 61 is a diagram schematically illustrating dNTP incorporation into a synthesized nucleic acid chain accompanied by the release of irrational pyrophosphate (PPi).

FIG. 62 is a diagram illustrating loading of beads into the microfluidic array of the present invention. FIG. 63 is a diagram illustrating loading of beads into the microfluidic array of the present invention. FIG. 64 is a diagram illustrating the loading of beads into the microfluidic array of the present invention. FIG. 65 is a diagram illustrating loading of beads into the microfluidic array of the present invention. FIG. 66 is a diagram illustrating loading of beads into the microfluidic array of the present invention. FIG. 67 is a diagram illustrating loading of beads into the microfluidic array of the present invention. FIG. 68 is a diagram illustrating loading of beads into the microfluidic array of the present invention. FIG. 69 is a diagram illustrating loading of beads into the microfluidic array of the present invention. FIG. 70 is a diagram illustrating the loading of beads into the microfluidic array of the present invention.

FIG. 71A shows a screen capture (see Tables 1 and 2) (left panel) showing the pixel of the signal that occurs after dATP (first) is added to template 4 resulting in a 4-base extension, and for the pixel indicated by the arrow. Plot of voltage versus frame (or time) (right panel).

FIG. 71B shows a screen capture (see Tables 1 and 2) showing the pixel of the signal that occurs after the next addition of dCTP resulting in 4-base extension in template 1 (left panel) and the voltage for the pixel indicated by the arrow. Plot of frame (or time) (right panel).

FIG. 71C shows a screen capture (see Tables 1 and 2) showing the pixel of the signal that occurs after dGTP is added resulting in stretching of templates 1, 2 and 4 (left panel) and the voltage versus voltage for the pixel indicated by the arrow. A frame (or time) plot (right panel).

FIG. 71D shows a screen capture (see Tables 1 and 2) showing the pixel of the signal that occurs after dTTP is added and runoff occurs in all 4 templates (due to the presence of all 4 dNTPs) (left panel) and FIG. 4 is a plot (right panel) of voltage versus frame (or time) for a pixel indicated by an arrow.

  The following is a more detailed description of related concepts or aspects, inventive methods and apparatus for large scale chemFET arrays for analyte measurement. It should be understood that the various concepts detailed above and below can be implemented in any of a number of ways, and that the disclosed concepts do not limit the specific methods of realization. . Examples of specific implementations and uses are provided primarily for illustrative purposes.

  Various inventive aspects according to the present disclosure relate to semiconductor-based / microfluidic hybrid systems that at least partially combine the capabilities of microelectronics and the biocompatibility of microfluidic systems. In the following examples, the microelectronic portion of the hybrid system is implemented by CMOS technology for illustrative purposes. However, the present disclosure is not intended to be limited in this regard, and other semiconductor-based technologies may be used to implement various aspects of the microelectronic portion of the system shown herein. Should be understood.

  One aspect shown herein relates to large arrays (eg, two-dimensional arrays) of chemically sensitive field effect transistors (chemFETs), where individual chemFET sensor elements or “pixels” of the array are in close proximity to the array. It is configured to detect analytical concentration changes in the host of chemical and / or biological processes that occur (chemical reactions, cell culture, neural activity, nucleic acid sequencing, etc.). Examples of chemFETs contemplated by the various aspects described in the more detailed description below include, but are not limited to, ion sensitive field effect transistors (ISFETs) and enzyme sensitive field effect transistors (ESFETs). In one exemplary implementation, one or more microfluidic structures are fabricated to provide retention and / or confinement of a chemical reaction that can generate an analyte of interest on a chemFET sensor array. For example, in one implementation, the microfluidic structure is configured as a “well” (eg, a small reaction chamber) that is disposed on one or more sensors of the array, with a predetermined well disposed thereon 1 Alternatively, two or more sensors may detect an analyte in a predetermined well and measure the concentration.

  In some embodiments, a chemFET array / microfluidic hybrid structure may be used for analysis of a solution / substance of interest that includes nucleic acids. For example, such structures can be used to process nucleic acids in a number of ways that utilize nucleic acid sequencing. In various aspects, such sequencing can include detecting single nucleotide polymorphisms in nucleic acid fragments, comparing nucleic acid expression profiles (compare nucleic acid expression profiles between two or more conditions, compare between disease and normal tissues, Comparing untreated and treated tissues of drugs, enzymes, irradiation or chemical treatment), haplotyping (compare genes or gene changes in each of the two alleles of human subjects), karyotyping (testing) A diagnostic comparison of one or more genes in the tissue, usually a gene from an embryo / fetus before conception to detect a congenital anomaly, with a “normal” karyotype subject), and For genotyping (compare one or more genes in a first individual of a species with the same gene in other individuals of the same species) It can be performed to determine the gender. However, while some examples of the concepts presented herein apply in the context of nucleic acid processing, it is understood that the application of the concepts relating to the chemFET sensor array presented herein is not limited to these examples. Should.

  FIG. 8 generally illustrates a nucleic acid processing system 1000 that includes a large-scale chemFET array according to one inventive aspect of the present disclosure. In the following description, the array of chemFET sensors shows, for illustrative purposes, an ISFET configured to detect hydrogen ion concentration. However, the present disclosure is not limited in this regard, and in any embodiment shown in the specification where an ISFET is used as an example, other types of chemFETs may be used in an alternative manner as well. It should be understood that In one aspect, the system 1000 includes a semiconductor / microfluidic hybrid structure 300 that includes an ISFET sensor array 100 and a microfluidic flow cell 200. In another aspect, the flow cell 200 includes nucleic acids disposed within the flow cell via controlled influx of a number of sequencing reagents 272 (eg, bases dATP, dCTP, dGTP, dTTP, and other reagents) into the flow cell. Configured to facilitate template sequencing. As shown in FIG. 8, the flow of sequencing reagent into the flow cell 200 can occur through one or more valves 270 and one or more pumps 274 controlled by a computer 260.

  In the system 1000 of FIG. 8, the ISFET sensor array 100 according to one aspect is a pH change that occurs in different parts of the flow cell 200 due to a chemical reaction between one or more bases comprising the sequencing reagent 272 and the nucleic acid template. To monitor. In other aspects described in more detail below, the FET sensor array may be specifically configured for the detection of other analytes that may provide relevant information about the chemical reaction of interest. Through the array controller 250 (also under the control of the computer 260), the ISFET array is controlled to acquire data relating to the analyte measurement, and the collected data generates valuable information relating to the processing of the nucleic acid template. To be processed. For example, in one implementation, the pH change is generally proportional to the number of a particular type of base (eg, one of dATP, dCTP, dGTP, dTTP) added to the nucleic acid template. Such a pH change can be represented by a change in the output voltage of one or more ISFETs in the array 100 in proximity to the reaction between a given base type and the template. Thus, the degree of voltage change for a given pixel in the array can be used to determine the number of a particular type of base added to the template placed in the flow cell on the given pixel.

  In one aspect, the flow cell 200 of the system 1000 shown in FIG. 8 may include a number of wells (not shown in FIG. 8) that are disposed on corresponding sensors of the ISFET array 100. Many techniques can be used to flow various treatment materials into the wells of such flow cells. For example, the nucleic acid to be sequenced can be loaded into a flow cell by first centrifuging “beads” containing a nucleic acid template into a well. Alternatively, such beads can enter the well by gravity. In another example, instead of using beads, the wells may be coated with a set of primer pairs and a nucleic acid template may be provided to the flow cell with an adapter that supplements the primer pair (immobilized material may be in the sensor array 100 or Can be added to individual dies as part of the chip package or just prior to nucleic acid processing). Other methods involving sol-gel may be used to immobilize nucleic acid templates near the surface of ISFET array 100.

  Once the nucleic acid template is loaded into each well of the flow cell 200, bridge amplification can then be performed in the wells, the product can be denatured and then sequenced by synthesis or ligation. Other amplification methods within the well (the state in which the product is captured in the well) are envisaged and include methods using rolling circle amplification or isothermal or non-isothermal amplification such as PCR. As shown in FIG. 8, a reagent containing a base may be flowed into the flow cell (eg, via a computer controlled valve 270 and pump 274) and diffused into the well, or by other methods such as ink jet into the flow cell. May be added. In yet another example, the flow cell 200 may not include wells, and the diffusion characteristics of the reagents can be utilized to limit crosstalk between the respective sensors of the ISFET array 100.

  That is, the flow cell 200 in the system of FIG. 8 can be configured in various ways to provide one or more analytes in close proximity to the ISFET array 100. For example, the nucleic acid template (DNA) may be directly attached or applied in appropriate proximity to one or more pixels of the sensor array 100, or a support material (eg, 1 Or two or more “beads”). Processing reagents (eg, enzymes) can also be placed directly on the sensor or on one or more solid supports proximate to the array, and the device can detect the enzyme with the sensor. It can be used without wells or beads for many biosensor applications that result in products (eg, changes in ion concentration).

  With respect to the ISFET array 100 of the system 1000 shown in FIG. 8, in one embodiment, the array 100 is designed and manufactured using a standard CMOS process (eg, 0.35 micrometer process, 0.18 micrometer process). Implemented as an integrated circuit, including all sensors and electronics required to monitor / measure one or more analytes. Referring back to FIG. 1, one or more reference electrodes 76 used in conjunction with the ISFET array 100 establish a baseline against which comparisons of analyte concentration changes in proximity to each ISFET in the array 100 can be compared. To do so, it may be placed in flow cell 200 (eg, placed in an unused cell of the flow cell) or exposed to another standard (eg, sequencing reagent 172). Reference electrode 76 can be electrically connected to array 100, array controller 250, or directly to computer 260 to facilitate analyte measurements based on voltage signals obtained from array 100. In some implementations, as described further below, the reference electrode may be connected to electrical ground or other predetermined potential to establish an electrical standard for ISFET output signal measurement, or the reference electrode The voltage may be measured with respect to ground.

  The ISFET array 100 is not limited to any particular size as a one or two dimensional array, but includes as few as 2 to 256 pixels (eg, a 2D implementation 16 × 16 pixels) or as many as 54 megapixels (2D Implementation 7400 × 7400 pixels), or many more, may be manufactured and used for various chemical / biological analysis purposes in accordance with the concepts disclosed herein. In one exemplary system embodiment shown in FIG. 8, each ISFET sensor in the array may be configured for detection of hydrogen ions. However, the present disclosure is not limited in this regard, and each sensor of the ISFET sensor array may be specially configured for sensitivity to other types of ion concentrations for a variety of applications. It should be understood (for example, materials sensitive to other ions such as sodium, silver, iron, boron, iodine, calcium or nitric acid are known).

  More generally, chemFET arrays according to various aspects of the present disclosure may be configured to detect one or more of the analyte / chemical types. In one aspect, one or more chemFETs of the array are configured to detect one or more analytes that represent one or more binding events (eg, related to a nucleic acid sequencing process), and other aspects Now, different chemFETs in a given array are configured to detect different analytes. For example, in one embodiment, one or more sensors (pixels) of the array may include a first type of chemFET configured to be chemically sensitive to the first analyte, and One or more other sensors in the array may include a second type of chemFET configured to be chemically sensitive to a second analyte that is different from the first analyte. In one exemplary implementation, the first analyte can represent a binding event for a nucleic acid sequencing process and the second analyte can represent a second binding event for a nucleic acid sequencing process. Of course, it should be understood that more than two different types of chemFETs may be used in any given array for detecting or measuring different types of analyte / binding events. In general, in any aspect of the sensor array presented herein, a given sensor array is “homogeneous” and detects and / or measures the same type of analyte (eg, pH or other ions). In order to detect and / or measure different types of analytes, the sensor array may include substantially similar or identical types of chemFETs, or the sensor array is “non-uniform”. It should be understood that a chemFET may be included. For the sake of brevity, the following are examples of ISFETs again in various aspects of the sensor array, but the present disclosure is not limited in this respect and several other options for analyte detection are described below. Shown in detail (eg, with respect to FIG. 11A).

  In a typical implementation based on 0.35 micrometer CMOS process technology (or CMOS process technology capable of smaller feature sizes), each pixel of ISFET array 100 includes an ISFET, with valid / select components, and approximately It may occupy an area on the array surface of 10 micrometers x 10 micrometers (ie, 100 square micrometers) or less. That is, an array having a pitch (pixel interval) on the order of 10 micrometers may be realized. An array pitch on the order of 10 micrometers or less using 0.35 micrometer CMOS process technology is a significant improvement in terms of size reduction over traditional ISFEST array manufacturing attempts that result in pixel sizes of at least 12 micrometers or larger. Bring.

  More specifically, based on the inventive concepts presented herein, in some embodiments shown in more detail below, an array pitch of about 9 micrometers, with row and column selection and bias / read electronics, An ISFET array containing more than 256,000 pixels (i.e. 512 x 512 arrays) can be fabricated on a 7 mm x 7 mm semiconductor die, and more than 4 million pixels (i.e. 4 x 2048 x 2048 arrays) It makes it possible to fabricate a similar sensor array containing more than megapixels) on a 21 mm × 21 mm semiconductor die. In another example, an array pitch of about 5 micrometers, an ISFET sensor array comprising about 1.55 megapixels (ie, 1348 × 1348 array), a 9 mm × 9 mm semiconductor with electronics. It makes it possible to fabricate into a die, and to produce an ISFET sensor array with 14 megapixels on a 22 mm × 22 mm semiconductor die with electronics. In yet another aspect, an ISFET sensor array with a pitch significantly below 5 micrometers using a CMOS manufacturing process (eg, 0.18 micrometers CMOS process technology) capable of feature sizes below 0.35 micrometers. Can be made (eg, 2.6 micrometer array pitch or 8 or 9 square micrometer pixel area), which provides an ultra-high density ISFET array. Of course, it should be understood that pixel sizes far beyond 10 micrometers (eg, on the order of about 20, 50, 100 micrometers or more) may be implemented in various aspects of the chemFET array according to the present disclosure. It is.

  In another aspect of the system shown in FIG. 8, one or more array controllers 250 are used to operate the ISFET array (eg, each pixel of the array to obtain an output signal indicative of the analyte measurement result). Select / enable). In various aspects, one or more components comprising one or more array controllers may be on the same integrated circuit (IC) chip as the array, but with different portions of the IC chip, along with the pixel elements of the array itself. Mounted off-chip. Regarding array control, the analog-to-digital conversion of the ISFET output signal is implemented on the same integrated circuit as the ISFET array, but is performed by a circuit located outside the sensor array area (analog-outside the sensor array area). By arranging the digital conversion circuit, it becomes possible to accommodate a smaller pitch and a large number of sensors thereby, and noise is also reduced). In various exemplary implementations, further described below, the analog-to-digital conversion may be 4 bits, 8 bits, 12 bits, 16 bits or other bit resolutions depending on the required signal dynamic range. Is possible.

  The general outline of the role of a chemFET (eg, ISFET) array 100 in an exemplary system 1000 for measuring one or more analytes for nucleic acid processing is as described above, but the following is a description of nucleic acid processing. Exemplary chemFET arrays of various inventive aspects of the present disclosure that can be used in a variety of applications, including but not limited to, are described in more detail. Again, for purposes of illustration, the chemFET arrays of the present disclosure will be described using typical examples of ISFETs, although other types of chemFET arrays can be used in alternative embodiments.

  As noted above, various inventive aspects of the present disclosure can significantly reduce pixel size and array pitch, thereby increasing the number of pixels in the ISFET array for a given semiconductor die size (ie, increasing pixel density). A special improvement was made to the Milgrew et al ISFET array design described above with respect to FIGS. 1-7, and other conventional ISFET array designs were similarly improved. In some implementations, the increase in pixel density, on the other hand, increases the signal-to-noise ratio (SNR) of the output signal corresponding to the respective measurement result for one or more chemical properties of one or more analytes and With an increase in the speed at which such output signals can be read from the array. In particular, we have relaxed the requirements for chemFET linearity and signal output corresponding to a more limited signal output / measurement range (eg, a pH range of about 7-9 rather than 1-14). ) Is recognized and understood that the complexity and size of individual pixels is greatly reduced, which facilitates the realization of ultra-large scale high density ISFET arrays.

To this end, Figure 9 illustrates the 102 _j is a row of ISFET arrays according to one inventive aspect of the present disclosure, it is considerably simplified in order to facilitate small pixel size ISFET pixel design here ing. Column 102 _j includes n pixels, with the first and last pixels shown in FIG. 9 as 105 ₁ and 105 _n . As described further below with respect to FIG. 13, a complete two-dimensional ISFET array based on the column design shown in FIG. 9 has m such columns 102 _j (j = 1, 2, 3, ..., m).

から

、負論理）の１つに応答する。列の全てのピクセルにあてはまる例として、ピクセル１０５_１を用いると、トランジスタスイッチＱ３は、ライン１１８_１を介して対応する行選択信号を受け取ると、ライン１１２_１を介して制御可能電源１０６_ｊと、ＩＳＦＥＴ１５０のソースとを接続する。トランジスタスイッチＱ２は、対応する行選択信号を受け取ると、ライン１１４_１を介して、ＩＳＦＥＴ１５０のソースと列バイアス／読取り回路１１０_ｊを接続する。ＩＳＦＥＴ１５０のドレインはライン１１６_１を介して直接バイアス／読取り回路と接続する。こうして、ピクセルあたり４つの信号、すなわちライン１１２_１、１１４_１、１１６_１および１１８_１のみが、ピクセル１０５_１の３つのコンポーネントを作動させるために必要とされる。ｍ列のアレイにおいて、所与の行選択信号は、各列の１つのピクセルに同時に印加される（例えば、各列の同じ位置）。 One aspect of the embodiment shown in FIG. 9, each pixel ₁₀₅ 1 to 105 _n of column 102 _j includes three components, namely, ISFET150 the (also referred Q1 both) and MOSFET switches Q2 and Q3. MOSFET switches Q2 and Q3 in order to enable or select a given pixel of the column 102 _j, both n row select signals (

From

, Negative logic). Examples that apply to all pixels of the column, the use of pixel 105 _1, transistor switch Q3 receives a corresponding row select signal via line 118 _1, a controllable power source 106 _j via line 112 _1, Connect to the source of ISFET 150. Transistor switch Q2 receives a corresponding row select signal via line 114 _1, connecting the source and column bias / readout circuitry 110 _j of ISFET150. Drain of ISFET150 is connected directly bias / readout circuitry via line 116 _1. Thus, only four signals per pixel, ie lines 112 ₁ , 114 ₁ , 116 ₁ and 118 _1, are required to activate the three components of pixel 105 ₁ . In an m-column array, a given row selection signal is applied simultaneously to one pixel in each column (eg, the same position in each column).

As shown in FIG. 9, the design of column 102 _j according to one aspect is based on a general principle similar to the design of Milgrew et al shown in FIG. 3 above. In particular, when the ISFET of each pixel is enabled, a constant drain current I _Dj and a constant drain-source voltage V _DSj are set in order to obtain the output signal V _Sj from the valid pixel according to the above equation (3). Is done. To this end, column 102 _j is connected to the analog circuit positive supply voltage VDDA and is responsive to the bias voltage VB1 by all the pixels in the column to provide a constant drain current I _Dj to the effective pixel ISFET. comprising a controllable power supply 106 _j being shared. In one aspect, power supply 106 _j is implemented as a current mirror that includes two long channel lengths and a high output impedance MOSFET. The column also includes a bias / read circuit 110 _j that is also shared by all the pixels in the column to provide a constant drain-source voltage to the effective pixel ISFET. The bias / read circuit 110 _j is based on the Kelvin bridge configuration, includes two operational amplifiers 107A (A1) and 107B (A2) arranged as buffer amplifiers, and is connected to the analog circuit positive power supply voltage VDDA and the analog ground power supply voltage VSSA. The The bias / read circuit also includes a controllable current sink 108 _j (similar to power supply 106 _j ) connected to analog ground VSSA and responsive to bias voltage VB2, and a diode connected MOSFET Q6. Bias voltages VB1 and VB2 are set / controlled in tandem to provide complementary source and sink currents. Voltage generated at both ends (-across) results diode connection MOSFETQ6 current consumption by the current sink 108 _j is by an operational amplifier, a constant drain - as the source voltage _{V DSJ,} as generated between the ISFET drain and source of effective pixels To be forced into.

The use of diode connected MOSFETQ6 the bias / readout circuitry within 110j in Figure 9, than the resistance _{R SDj} shown in Milgrew et al design shown in FIG. 3, provides significant advantages to the CMOS manufacturing process. Specifically, matching resistors can generally be manufactured with tolerances on the order of ± 20%, while MOSFET matching in CMOS fabrication is on the order of ± 1% or better. The degree to which a component responsible for providing a constant ISFET drain-source voltage V _DSj can be matched from column to column has a significant impact on measurement accuracy (eg, offset) between columns. Therefore, using MOSFET Q6 rather than a resistor significantly mitigates the measurement offset between columns. Further, while the thermal drift characteristics of the resistor and ISFET are quite different, the thermal drift characteristics of the MOSFET and ISFET are not substantially the same, but are substantially similar, and the thermal drift in the MOSFET Q6 is substantially equivalent to the thermal draft from the ISFET. And provides better measurement stability against changes in array temperature.

In FIG. 9, column bias / read circuit 110 _j also includes a sampling / hold and buffer circuit that provides an output signal V _COLj from the column. In particular, after one pixel of pixels 105 ₁ to 105 _n is enabled or selected via transistors Q2 and Q3 in each pixel, the output of amplifier 107A (A1), ie, V _Sj, is the column sampling and holding capacitor It is stored in C _Sh via a switch operation in response to the column sampling and hold signal COL SH. Examples of suitable capacitance for sampling and holding capacitors include, but are not limited to, a range of about 500 fF to 2 pF. The sampled voltage is buffered via column output buffer amplifier 111 _j (BUF) and provided as column output signal V _COLj . As shown in FIG. 9, the reference voltage VREF is applied to the buffer amplifier 111 _j via a switch CAL responsive to the control signal to facilitate the characterization of the non-uniformity between columns by the buffer amplifier 111 _j and thus read. Allows later data modification.

Figure 9A illustrates an exemplary circuit diagram for one of the amplifiers 107A of the bias / readout circuitry 110 _j, (amplifier 107B is implemented similarly), FIG. 9B, 107A and 107B of the amplifier bias and bandwidth It is a graph of. As shown in FIG. 9A, the amplifier 107A is configured as a unity gain buffer using a plurality of current mirror arrangements based on nine MOSFETs (M1 to M9), where the input and output of the amplifier are generally IN + And VOUT respectively. The bias voltage VB4 (indicating the corresponding bias current) controls the transimpedance of the amplifier and also serves as bandwidth control (ie, the bandwidth increases with increasing current). Referring again to FIG. 9, with the sampling and holding capacitor C _sh , the output of amplifier 107A substantially drives the filter when the sampling and holding switch is closed. Thus, in order to achieve a fairly high data rate, the bias voltage VB4 can be adjusted to increase the higher bias current and amplifier bandwidth. From FIG. 9B, it can be seen that in some exemplary implementations, amplifier bandwidths of at least 40 MHz and significantly greater can be achieved. In some implementations, amplifier bandwidths as high as 100 MHz may be suitable to facilitate high data acquisition rates and relatively low pixel samples or dwell times (eg, on the order of 10-20 microseconds).

In another aspect of the embodiment shown in FIG. 9, unlike the Millgrew et al pixel design shown in FIG. 3, pixels 105 ₁ to 105 _n can be any transmission gate that requires both n-channel and p-channel. Or other drives. In particular, pixels 105 ₁ to 105 _n of this embodiment include the same type of FET device (ie, n-channel only or p-channel only). For illustrative purposes, 105 ₁ through 105 _n shown in FIG. 9 are shown as including only p-channel components, ie, two p-channel MOSFETs Q 2 and Q 3 and p-channel ISFET 150. By not using a transfer gate for connecting the source and the bias / readout circuitry 110 _j of ISFET, some of the dynamic range for ISFET output signal may be sacrificed. However, we allow some output signal dynamic range to be eliminated (and thereby limit the measurement range of a given chemical property such as pH. Recognizes and understands that the requirement of different types of FET devices in each pixel (both n-channel and p-channel) is eliminated and the number of pixel components is reduced. This facilitates pixel size reduction, as shown in more detail below with respect to FIGS. Thus, in one aspect, there is a beneficial tradeoff between reduced dynamic range and smaller pixel size.

In yet another aspect of the embodiment shown in FIG. 9, unlike the Millgrew et al pixel design, the ISFET 150 of each pixel 105 ₁ to 105 _n has no body connection to the source (ie, the ISFET's in operation). There is no electrical conductor connecting the body connection and the source of the ISFET that forces the body connection and source to the same potential). On the contrary, the body connections of all the ISFETs in the array are coupled together and lead to the body bias voltage V _BODY . Although not clearly shown in FIG. 9, the body connections to MOSFETs Q2 and Q3 are similarly not connected to their respective sources, but are connected to the body bias voltage V _BODY . In one exemplary implementation based on pixels that all have p-channel components, the body bias voltage V _BODY can be connected to the maximum voltage applicable to the array (eg, VDDA), as described below with respect to FIG. .

Because the body connection of each ISFET is not connected to its source, the non-zero source-body voltage V _SB causes the “body effect” described above with respect to FIG. 1 and affects the threshold voltage V _TH of the ISFET in a non-linear relationship. (Thus, according to equation (3), the measurement results of chemical properties such as pH may be affected). However, we recognize that by focusing on the reduced ISFET output signal dynamic range, the body effects that can be generated from a non-zero source-body voltage V _{SB in} the ISFET can be relatively small, I understand. Thus, some measurement non-linearities that may occur with reduced dynamic range may be ignored as insignificant or may be corrected to account for (eg, array calibration as described in more detail below with respect to FIG. 17). And through data processing technology). The inventors also allow each of the FETs that make up a pixel to share a common body connection, as shown below with respect to FIGS. Further, the pixel size can be easily reduced. Thus, in another aspect, there is a beneficial tradeoff between reduced linearity and pixel size.

Figure 10 illustrates the plan view of the chip layout design pixel 105 ₁ shown in FIG. 9, according to one aspect of the present disclosure. FIG. 11A illustrates a composite cross-sectional view along the line I--I of the pixel shown in FIG. 10, with II--II and III--III lines in the right half of FIG. 12A to 12L show plan views of the interlayer view shown in FIG. 11A (each image in FIGS. 12A to 12L is the pixel layout design shown in FIG. 10). Are stacked on top of each other to create a). In one exemplary implementation, the pixel design shown in FIGS. 10-12 uses a standard 4-metal, 2-poly, 0.35 micrometer CMOS process and has a dimension “about 9 micrometers” as shown in FIG. This can be achieved by providing a geometrically square pixel having a dimension “f” corresponding to an e ”and an ISFET sensitive portion of about 7 micrometers.

In the plan view of FIG. 10, ISFET 150 (denoted as Q1 in FIG. 10) generally occupies the right center of the pixel example, and the gate, source, and drain positions are denoted by Q1 _G , Q1 _S, and Q1 _D. . MOSFETs Q2 and Q3 generally occupy the left center of the example pixel, and the gate and source of MOSFET Q2 are represented by Q2 _G and Q2 _S , and the gate and source of MOSFET Q3 are represented by Q3 _G and Q3 _S. In one aspect of the layout shown in FIG. 10, MOSFET Q2 and Q3 share a drain represented by Q2 / _{3 D.} In another aspect, from the top view of FIG. 10, the ISFET is formed such that its channel is along the first axis of the pixel (eg, parallel to lines I-I), while MOSFETs Q2 and Q3 are channels whose pixels are pixels. It can be generally seen that it is formed along a second axis perpendicular to the first axis. FIG. 10 also shows the four lines needed to operate the pixel, namely the line 112 ₁ connected to the source of Q3, the line 114 ₁ connected to the source of Q2, and the drain of the ISFET. The row selection line 118 ₁ connected to the line 116 ₁ to be connected and the gates of Q2 and Q3 is shown. Referring to FIG. 9, all pixels in a given column share lines 112, 114 and 116 (running pixels vertically in FIG. 10), and all pixels in a given row use line 118. I can understand that it is shared. Thus, it can be seen that in the layout shown in FIG. 10 based on the pixel design of FIG. 9, only four metal lines are needed to traverse each pixel.

  Referring now to the cross-sectional view of FIG. 11A, heavily doped p − -type regions 156 and 158 (located along line I—I in FIG. 10) are located in the n-well 154 in the ISFET source (S) and Between the drain (D) is an n-well region 160 with an ISFET polysilicon gate 164 and an ISFET p-channel formed below the gate oxide. In one aspect of the inventive embodiment shown in FIGS. 10 and 11, all FET components of pixel 1051 are fabricated as p-channel FETs in a single n-type well 154 formed in p-type semiconductor substrate 152. . This is possible because, unlike the Milgrew et al design, 1) no transmission gate is required in the pixel and 2) the ISFET source is not connected to the n-well body connection. More specifically, as shown in FIG. 10, the highly doped n-type region 162 provides a body connection (B) to the n-well 154, and the body connection B is connected to the metal conductor 332 around the boundary of the pixel 1051. Connecting. However, the body connection is not directly electrically connected to the source region 156 of the ISFET (ie, there is no electrical conductor connecting the body connection and the source that forces the body connection and the source to the same potential during operation), The body connection does not electrically connect to the gate, source and drain of any component in the pixel. Thus, the other p-channel FET components of the pixel, Q2 and Q3, are fabricated in the same n-well 154.

  The composite cross-sectional view of FIG. 11A also shows a highly doped p-type region 159 (located along line I--I in FIG. 10), corresponding to the shared drain (D) of MOSFETs Q2 and Q3. For illustrative purposes, the polysilicon gate 166 of MOSFET Q3 is also shown in FIG. 11A, but this gate is not located along line I--I in FIG. 10, but rather is a cross-sectional view along line I--I. Located on the back of the. However, for simplicity, the sources of MOSFETs Q2 and Q3 shown in FIG. 10 and also the gate of Q2 are located along the same axis as the shared drain (ie, perpendicular to the plane of the figure) Not shown in FIG. 11A (if shown in FIG. 11A, these elements would overcomplicate the composite cross-sectional view of FIG. 11A).

Above the substrate, gate oxide and polysilicon layers shown in FIG. 11A, a number of additional layers are provided to establish electrical connections with various pixel components, which are conductive vias. Including alternating metal layers and oxide layers. According to the 4-metal CMOS process, these layers are labeled “contact”, “metal 1”, “via 1”, “metal 2”, “via 2”, “metal 3”, “via 3” in FIG. 11A. And “Metal 4”. To make the ISFET electrical connection more comprehensible, the composite cross-sectional view of FIG. 11A shows additional elements of pixel fabrication between the right side lines II--II and lines III--III of the plan view of FIG. With respect to the electrical connection of the ISFET, the top metal layer 304 corresponds to the ISFET detection region 178, on which the analyte-sensitive passivation layer 172 is disposed. The top metal layer 304 is an ISFET “floating” with ISFET polysilicon gate 164 and intervening conductors 306, 308, 312, 316, 320, 326 and 338 in a similar manner associated with the conventional ISFET design shown in FIG. A “gate” structure 170 is formed. Electrical connection between the ISFET drain is provided by conductors 340,328,318,314 and 310 for connecting ₁ to the line 116. The ISFET source is connected to the shared drains of MOSFETs Q2 and Q3 via conductors 334 and 336 and 324 (these are located on lines I--I in FIG. 10). The body connection 162 to the n-well 154 is electrically connected to the metal conductor 322 around the pixel on the “metal 1” layer via conductors 330 and 322.

  As described above, FIGS. 12A to 12L are plan views of the interlayer diagram shown in FIG. 11A (each image of FIGS. 12A to 12L is one by one to create the pixel layout design shown in FIG. 10). Layered on top). In FIG. 12, the correspondence between the plan view with letters of each layer and the cross-sectional view of FIG. 11A is as follows: A) n-type well 154, B) ion implantation part (implant), C) diffusion part (diffusion) D) polysilicon gates 164 (ISFET) and 166 (MOSFETs Q2 and Q3), E) contacts, F) metal 1, G) via 1, H) metal 2, I) via 2, J) metal 3, K) via 3, L) Metal 4 (top electrode connected to ISFET gate). The various reference numbers shown in FIGS. 12A to 12L are the same as those shown in the composite cross-sectional view of FIG. 11A.

  Thus, the pixel chip layout design shown in FIGS. 10, 11 and 12A through 12L according to one aspect is the same type of FET device used for all components of the pixel, with all components mounted in a single well. It illustrates what can be done. This dramatically reduces the area required for the pixel, which facilitates increasing the pixel density in a given area.

In one exemplary implementation, the ISFET gate oxide 165 is fabricated with a thickness on the order of about 75 Angstroms and the capacitance C _ox per unit area of the gate oxide is increased to 4.5 fF / μm ^2. sell. Further, the polysilicon gate 164 may be fabricated with dimensions corresponding to a channel width W of 1.2 μm and a channel length L of 0.35 to 0.6 μm (ie, W / L is about 2 to 3.5). The doping of region 160 may be selected such that the carrier mobility of the p-channel is 190 cm ² / V · s (ie 1.9E10 μm ² / V · s). From equation (2) above, this places the ISFET transconductance parameter β on the order of about 170-300 μA / V ² . In another aspect of this exemplary implementation, the analog power supply VDDA is 3.3 volts, and VB1 and VB2 are biased to provide a constant ISFET drain current I _Dj on the order of 5 μA (some In implementation, VB1 and VB2 may be set to provide a drain current of about 1 μA to 20 μA). Further, MOSFET Q6 (see bias / readout circuitry 110 _j in FIG. 9) is given _{I Dj} is 5 .mu.A, the voltage across 800 mV _(i.e., V DSj = 800 mV) and so that, the channel width and length of the The ratio (for example, W / L is about 50) is adjusted. From equation (3), based on typical parameters, this provides a pixel output voltage with a width of about 0.5-2.5V for changes in ISFET threshold voltage over a range of about 0-2V. .

With respect to the analyte-sensitive passivation layer 172 shown in FIG. 11A, in a typical CMOS implementation, the passivation layer may be significantly sensitive to hydrogen ion concentration and may be silicon nitride (Si ₃ N ₄ ) and / or acid. Silicon nitride may be included. In conventional CMOS processes, the passivation layer may be formed by continuous deposition of one or more of these materials, and generally treats the device to protect against contamination and improve electrical stability. It may be used for coating. The physical properties of silicon nitride and silicon oxynitride significantly barrier protection from scratches and water and salt diffusion, where passivation layers containing these materials can corrode device metallization and / or destabilize device operation. Is. A passivation layer comprising silicon nitride and / or silicon oxynitride provides ion sensitivity in an ISFET device, which includes surface groups that can release or accept protons by contacting the analysis solution. This changes the surface potential and device threshold voltage V _{TH as} described above with respect to FIG.

  In a CMOS process that includes aluminum (having a melting point of about 650 degrees Celsius) as a metal, a silicon nitride and / or silicon oxynitride passivation layer is deposited by plasma enhanced chemical vapor deposition (PECVD), which is The glow discharge of 350 degrees ionizes the constituent gases that form silicon nitride or silicon oxynitride and reacts on the wafer surface to create active species that form a laminate of the respective materials. In one exemplary process, a passivation layer having a thickness on the order of about 1.0-1.5 μm is first deposited with a thin layer of silicon oxynitride (on the order of 0.2-0.4 μm), followed by A slightly thick silicon oxynitride may be deposited (on the order of 0.5 μm) and finally silicon nitride (on the order of 0.5 μm) may be deposited. Since the PECVD process involves a low deposition temperature, aluminum metallization is not adversely affected.

However, the inventors have found that the low temperature PECVD process provides a suitable passivation layer for conventional CMOS devices, while the low temperature process generally results in a low density and somewhat porous passivation layer, In that case, it is recognized and understood to adversely affect the stability of the ISFET threshold voltage. In particular, during the operation of the ISFET device, low density porous passivation layer begins to absorb the analyte or other substances in the solution with time saturation, the ISFET threshold voltage V _TH, undesirable to difficult accurate measurement Causes time-dependent drift.

  In view of the above, in one embodiment, a CMOS process using tungsten instead of aluminum may be used to fabricate an ISFET array according to the present disclosure. Due to the high melting point of tungsten (over 3400 degrees Celsius), a high temperature low pressure chemical vapor deposition (LPCVD) process (eg, 700-800 degrees Celsius) can be used for silicon nitride or silicon oxynitride passivation layers. The LPCVD process generally results in a denser and less porous film in the passivation layer, thereby mitigating the adverse effects of ion absorption from the analytical solution causing ISFET threshold voltage drift.

  In yet another embodiment used to fabricate an ISFET array according to the present disclosure using an aluminum-based CMOS process, the passivation layer 172 shown in FIG. 11A can be used for other additional deposition beyond the range typically used in conventional CMOS processes. And / or may include substances. For example, the passivation layer 172 may initially include the low temperature plasma enhanced chemical vapor deposition (PECVD) of silicon nitride and / or silicon oxynitride as described above. For purposes of this description, this conventional deposition is shown in FIG. 11A as the first portion 172A of the passivation layer 172. FIG. In one embodiment, following the first portion 172A, one or more passivation materials may increase the overall density of the passivation layer 172 and at least a second portion to reduce porosity (and adsorption by the porous). Vapor deposited to form 172B. Although one additional portion 172B is shown primarily in FIG. 11A for purposes of illustration, the present disclosure is not limited in this regard, and the entire passivation layer 172 may include two or more components, where It should be understood that each portion may include one or more layers / depositions that are the same or different, and that each portion may be configured similarly or differently.

Examples of materials suitable as the second portion 172B (or other additional portion) of the passivation layer 172 that is particularly sensitive to hydrogen ions include, but are not limited to, silicon nitride, silicon oxynitride, aluminum oxide (Al ₂ O ₃ ), tantalum oxide (Ta ₃ O ₅ ), tin oxide (SnO ₂ ) and silicon dioxide (SiO ₂ ). In one aspect, the second portion 172B (or other additional portion) can be a variety of relatively low temperature processes including, but not limited to, RF sputtering, DC magnetron sputtering, thermal or electron beam deposition, and ion-assisted deposition. Vapor deposition may be performed. In another aspect, a pre-sputtering etching process may be used to remove native oxide on the first portion 172A (or elevated temperature in a hydrogen environment, etc.) prior to the deposition of the second portion 172B. May be used to remove the native oxide on the first portion 172A). Further, in another aspect, the thickness of the second portion 172B may be on the order of about 0.04 μm to 0.06 μm (400 to 600 angstroms), and the thickness of the first portion may be as described above. The order may be 1.0 to 1.5 μm. In some exemplary implementations, the first portion 172A may include multiple layers of silicon nitride and silicon oxynitride having a thickness of 1.0-1.5 μm all inclusive and the second portion 172B may comprise a single layer of about 400-600 Angstroms of aluminum oxide or tantalum oxide. It should also be understood that the above-described exemplary thicknesses are primarily intended to be exemplary and the present disclosure is not limited in these respects.

  As such, the chemFET arrays shown herein can be used to detect a variety of analytes, thereby monitoring a variety of reactions and / or interactions. The emphasis on hydrogen ion detection (in the form of pH changes) is for convenience and brevity, and other analytes (including other ions) replace hydrogen in these descriptions. sell.

  The chemFETs shown herein include ISFETs and can detect any analyte that can itself cause a change in the electric field. The analyte need not be charged in order to be detected by the sensor. For example, depending on the embodiment, the analyte may be positively charged (ie, cation), negatively charged (ie, anion), or zwitterion (ie, having two equal or opposite charges). May be but not totally neutral) and polar yet neutral. This list is not intended to cover other analyte categories as well as the types in each category that can be readily predicted by one of ordinary skill in the art based on this disclosure.

  In the broadest scope of the present invention, the passivation layer may or may not be coated, and the analyte may or may not interact with the passivation layer. As an example, the passivation layer may consist of silicon nitride and the analyte may be something that is not hydrogen ions. As a special example, the passivation layer may consist of silicon nitride and the analyte may be inorganic pyrophosphate (PPi). In these examples, PPi is detected directly (ie, when no PPi receptor is present directly or indirectly in the passivation layer).

  If the analyte to be detected is hydrogen ions (or alternatively hydroxide), it is preferred to use a weak buffer so that any change in ionic species is detected in the passivation layer. If the analyte to be detected is not a hydrogen ion (or hydroxide), there is some possibility that the pH of the solution will change during the reaction or detection stage, so that the pH change does not interfere with the detection of the analyte. It is preferred to use a strong buffer. A buffer is an ionic molecule that resists changes in pH. The buffer neutralizes the acid or base added or generated to the solution, and as a result, the pH of the solution does not change. It should be understood that the buffer is reasonably provided to have the desired range of pKa. Suitable buffers are those that function in the pH range 6-9 and more preferably 6.5-8.5. The strength of the buffer is a relative condition because it depends on the nature, strength and concentration of the acid or base added or produced in the solution of interest. Weak buffers are approximately at least +/− 0.005, 0.01, 0.015, 0.02, 0.03, 0.04, 0.05, 0.10, 0.15, 0.20, 0 A buffer that allows detection of pH changes of .25, 0.30, 0.35, 0.45, 0.50 or more (and therefore cannot be controlled otherwise) . In some embodiments, the pH change is on the order of about 0.005 (eg, per nucleotide incorporation), preferably an increase in pH. Strong buffers are at least about +/− 0.005, 0.01, 0.015, 0.02, 0.03, 0.04, 0.05, 0.10, 0.15, 0.20, 0 Control pH changes of .25, 0.30, 0.35, 0.45, 0.50 or more. The strength of the buffer can be evaluated by changing the concentration of the buffer species itself. Thus, a low concentration buffer can be a low intensity buffer. Examples include less than 1 mM (eg, 50-100 μM) buffer species. A non-limiting example of a weak buffer suitable for a sequencing reaction in which the pH change shown herein is read information is 0.1 mM Tris or Tricine. Examples of suitable weak buffers are given in the examples and are known to those skilled in the art. A higher concentration of buffer can be a higher intensity buffer. Examples include those with 1-25 mM buffer species. Non-limiting examples of strong buffers suitable for sequencing reactions where PPi is directly read as shown herein are 1.5 or 25 mM (or greater) Tris or Tricine. One skilled in the art will be able to determine suitable buffers and detection methods for the reactions encompassed by the present invention.

  In some embodiments, the passivation layer and / or molecules coated thereon determine analyte specificity for array reading.

Detection of hydrogen ions (in the form of pH) includes silicon nitride (Si ₃ N ₄ ), silicon oxynitride (Si ₂ N ₂ O), silicon dioxide (SiO ₂ ), aluminum oxide (Al ₂ O ₃ ), pentoxide This is done using a passivation layer made of tantalum (Ta ₂ O ₅ ), tin oxide or stannic oxide.

  The passivation layer directly detects other ionic species including, but not limited to calcium, potassium, sodium, iodide, magnesium, chloride, lithium, lead, silver, cadmium, nitric acid, phosphoric acid, dihydrogen phosphate, etc. You can also

  In some embodiments, the passivation layer is coated with a receptor for the analyte of interest. The receptor selectively binds to the analyte of interest. As used herein, a receptor that selectively binds to an analyte is a molecule that selectively binds to the analyte (ie, the binding affinity for that analyte is greater than the binding affinity for other analytes). large). The binding affinity for the analyte of interest is 2 times, 3 times, 4 times, 5 times, 6 times, 7 times, 8 times, 9 times, 10 times, 15 times, 20 times that of other analytes. 25 times, 30 times, 40 times, 50 times, 100 times or more. In addition to its relative binding affinity, the receptor must also have an absolute binding affinity that binds with sufficient efficiency to the analyte of interest (ie it must have sufficient selectivity). Receptors having a binding affinity in the picomolar to micromolar range are suitable. Such interaction is preferably reversible.

  The receptor may be of any nature (eg, chemicals, nucleic acids, peptides, lipids, mixtures thereof, etc.). Analytes can also be of any nature and can be provided where a selectively binding receptor is present, and can also be seen in some embodiments. However, it should be understood that the present invention is intended to detect an analyte even in the absence of a receptor. One example of this is PPi or Pi detection by the passivation layer in the absence of PPi and Pi receptors.

  In one aspect, the invention contemplates a receptor that is an ionophore. As used herein, an ionophore is a molecule that selectively binds to an ionic species, here an anion or cation. In the context of the present invention, an ionophore is a receptor or ion that binds to an analyte. The ionophore of the present invention includes a carrier ionophore (ie, a low fat-soluble molecule that binds to a specific ion) that is recognized in the art as derived from microorganisms. Various ionophores are commercially available, such as from Calbiochem.

  The detection of some ions is achieved through the use of the passivation layer itself or a receptor coated on the passivation layer. For example, potassium can be selectively detected using polysiloxane, valinomycin or salinomycin, sodium can be selectively detected using monensin, nystatin or SQI-Pr, and calcium can be ionomycin, calcium. It can be selectively detected using mycin (A23187) or CA1001 (ETH1001).

  Receptors that can bind two or more ions can also be used in some examples. For example, beauvericin can be used for detection of calcium and / or barium ions, nigericin can be used for detection of potassium, hydrogen and / or lead ions, and gramicidin can be detected for hydrogen, sodium and / or potassium ions. Can be used. One skilled in the art will recognize that these compounds can be used in applications where a single ion specificity is not required or where other ions to which the compound binds are difficult (or impossible) to exist or are not generated. .

In other embodiments, including but not limited to nucleic acid sequencing, receptors that selectively bind to inorganic pyrophosphate (PPi) can be used. Examples of PPi receptors include the compounds shown in FIG. 11B (compounds 1-10). Compound 1 is described in Angew Chem Int Ed 2004 43: 4777-4780 and US 2005/0119497 A1, and is referred to as p-naphthyl-bis [(bis (2-pyridylmethyl) amino) methyl] phenol. Compound 2 is described in J Am Chem Soc 2003 125: 7752-7753 and US 2005/0119497 A1, p- (p-nitrophenylazo) -bis [(bis (2-pyridylmethyl-1) amino) methyl] It is called phenol (or its binuclear Zn complex). Compound 3 is described in Sensors and Actuators B 1995 29: 324-327. Compound 4 is described in Angew Chem Int Ed 2002 41 (20): 3811-3814. Compound 5 is described in WO 2007/002204, in which bis ^{-Zn 2+} - is referred to as dipicolylamine ^(Zn 2+ -DPA). The synthetic route for compounds 1 and 2 is shown by US 2005/0119497 A1. A typical synthesis of compound 4 is shown in FIG. 11C. Addition of compound 10 to the metal oxide is shown in FIG. 11E.

  As another example, receptors for neurotoxins are described in Simonian Electroanalysis 2004, 16: 1896-1906.

  The receptor can be covalently or non-covalently attached to the passivation layer. Covalent attachment to receptor passivation can be direct or indirect (eg, via a linker). FIG. 11D illustrates the use of silanol chemistry to covalently bond the receptor and the passivation layer. The receptor may be immobilized on the passivation layer using, for example, a primary aliphatic amine (lower left panel) or aryl isothiocyanate (lower right panel). In these and other embodiments, the passivation layer, which can itself consist of silicon nitride, aluminum oxide, silicon oxide, tantalum pentoxide, etc., is bonded to the silanized layer via its active surface groups. For details on silanol chemistry for covalent attachment of FET surfaces, at least the following publications may be referenced: For silicon nitride, see Sensors and Actuators B 1995 29: 324-327, Jpn J Appl Phys 1999 38: 3912-3917 and Langmuir 2005 21: 395-402. For information on silicon oxide, see Protein Sci 1995 4: 2532-2544 and Am Biotechnol Lab 2002 20 (7): 16-18. For aluminum oxide, see Colloids and Surfaces 1992 63: 1-9, Sensors and Accuators B 2003 89: 40-47 and Bioconjugate Chem 1997 8: 424-433. The receptor then binds to the active group of the silanized layer. As shown in FIG. 11D, this latter linkage can occur directly or indirectly using a bifunctional linker. A bifunctional linker is a compound having at least two reactive groups to which two objects can be attached. In some examples, the reactive group is located at the opposite end of the linker. In some embodiments, the bifunctional linker is a universal bifunctional linker as shown in FIG. 11D. Universal bifunctional linkers can be used to attach a variety of objects. It should be understood that the chemicals shown in FIG. 11D are meant to be illustrative and not limiting.

  The bifunctional linker may be a homobifunctional linker or a heterobifunctional linker depending on the nature of the molecule to be attached. A homobifunctional linker has two identical reactive groups. Heterobifunctional linkers have two different reactive groups. Various commercially available linkers react with one or more of the following groups: primary amines, secondary amines, sulfhydryls, carboxyls, carbonyls and carbohydrates. Examples of amine-specific linkers include bis (sulfosuccinimidyl) suberate, bis [2- (succinimidooxycarbonyloxy) ethyl] sulfone, disuccinimidyl suberate, disuccinimidyl tartrate, dimethyl adipate 2HCl, dimethyl pimelimate, 2HCl, dimethyl suberimidate, 2HCl, and ethylene glycol bis [[succinic acid] succinimidyl]. The linker that reacts with the sulfhydryl group is bismaleimidohexane, 1,4-di- [3 ′-(2′-pyridyldithio) -propionamide)] butane, 1- [p-azidosalicylamido] -4- [iodo. Acetamido] butane, and N- [4- (p-azidosalicylamido) butyl] -3 ′-[2′-pyridyldithio] propionamide. Linkers that preferentially react with carbohydrates include azidobenzoylhydrazine. Linkers that selectively react with carboxyl groups include 4- [p-azidosalicylamido] butylamine.

  Heterobifunctional crosslinkers that react with amines and sulfhydryls include N-succinimidyl-3- [2-pyridyldithio] propionate, succinimidyl [4-iodoacetyl] aminobenzoate, succinimidyl 4- [N-maleimidomethyl] cyclohexane- 1-carboxylate, m-maleimidobenzoyl-N-hydroxysuccinimide ester, sulfosuccinimidyl 6- [3- [2-pyridyldithio] propionamido] hexanoate and sulfosuccinimidyl 4- [N-maleimidomethyl] Cyclohexane-1-carboxylate is included. Heterobifunctional linkers that react with carboxyl and amino groups include 1-ethyl-3- [3-dimethylaminopropyl] -carbodiimide hydrochloride. Heterobifunctional linkers that react with carbohydrates and sulfhydryls include 4- [N-maleimidomethyl] -cyclohexane-1-carboxyl hydrazide 2HCl, 4- (4-N-maleimidophenyl) -butyric acid hydrazide 2HCl and 3- [2-Pyridyldithio] propionyl hydrazide is included.

  Alternatively, the receptor may be non-covalently coated on the passivation layer. Non-covalent deposition of the receptor on the passivation layer may involve the use of a polymer matrix. The polymer may be naturally occurring or non-naturally occurring and is not limited to nucleic acids (eg, DNA, RNA, PNA, LNA, etc., or mimetics, derivatives or combinations thereof). , Amino acids (eg, peptides, proteins (raw or denatured), etc., or mimetics, derivatives or combinations thereof, lipids, polysaccharides, and functional block copolymers. Or you may be taken in inside.

  Alternatively, the receptor may be covalently linked or crosslinked to the polymer (eg, “grafted” onto the functional polymer).

  An example of a suitable peptide polymer is polylysine (eg, poly-L-lysine). Examples of other polymers include polyethylene glycol (PEG), polyamide, polycarbonate, polyalkylene, polyalkylene glycol, polyalkylene oxide, polyalkylene terephthalate, polyvinyl alcohol, polyvinyl ether, polyvinyl ester, halogenated polyvinyl, polyvinyl pyrrolidone, poly Glycolide, polysiloxane, polyurethane, alkylcellulose, hydroxyalkylcellulose, cellulose ether, cellulose ester, nitrocellulose, block copolymer containing acrylic acid polymer and methacrylic acid ester, methylcellulose, ethylcellulose, hydroxypropylcellulose, hydroxypropylmethylcellulose, hydroxybutyl Methyl cellulose, cellulose acetate, pro Cellulose onate, cellulose acetate butyrate, cellulose acetate phthalate, carboxyl ethyl cellulose, cellulose triacetate, cellulose sulfate sodium salt, poly (methyl methacrylate), poly (ethyl methacrylate), poly (butyl methacrylate), poly (methacrylic acid) Isobutyl), poly (hexyl methacrylate), poly (isodecyl methacrylate), poly (lauryl methacrylate), poly (phenyl methacrylate), poly (methyl acrylate), poly (isopropyl acrylate), poly (isobutyl acrylate) ), Poly (octadecyl acrylate), polyethylene, polypropylene, poly (ethylene glycol), poly (ethylene oxide), poly (ethylene terephthalate), poly (vinyl alcohol), polyvinyl acetate, polyvinyl chloride, Restyrene, polyhyaluronic acid, casein, gelatin, glutin, polyanhydride, polyacrylic acid, alginate, chitosan, poly (methyl methacrylate), poly (ethyl methacrylate), poly (butyl methacrylate), poly (isobutyl methacrylate) ), Poly (hexyl methacrylate), poly (isodecyl methacrylate), poly (lauryl methacrylate), poly (phenyl methacrylate), poly (methyl acrylate), poly (isopropyl acrylate), poly (isobutyl acrylate) , And poly (octadecyl acrylate), poly (lactide-glycolide), copolyoxalate, polycaprolactone, polyesteramide, polyorthoester, polyhydroxybutyric acid, polyanhydride, poly (styrene-b-isobutylene-b-styrene (SIBS) block copolymers, vinyl ethylene acetate, poly (meth) acrylic acid, polymers of lactic acid and glycolic acid, polyanhydrides, poly (ortho) esters, polyurethanes, poly (butyric acid), poly (valeric acid), and poly (Lactide-cocaprolactone), and natural polymers such as alginate and polysaccharides including dextran and cellulose, collagen, albumin and other hydrophilic proteins, zein and other prolamins and hydrophobic proteins, copolymers and mixtures thereof, and chemical groups These include chemical derivatives thereof, including, for example, substitutions and / or additions such as alkyl, alkylene, hydroxylation, oxidation, and modifications that can be routinely performed by those skilled in the art.

  Another issue related to ISFET threshold voltage stability and / or predictability is that various process activities during or after array fabrication (eg, post processing such as plasma metal etching, wafer cleaning, dicing, packaging, handling, etc.) ) As a result of trap charges that can accumulate in the metal layer of the CMOS device. In particular, referring to FIG. 11A, in some examples, trapped charges may be generated by various conductors 304, 306, 308, 312, 316, 320, 326, 338 and one or more ISFET floating gate structures 170. 164 can be accumulated on 164. This phenomenon is also called “antenna effect” in related literature.

  One opportunity to accumulate trapped charges includes plasma etching of the top metal layer 304. We have the opportunity to accumulate charge on one or more conductors of a floating gate structure, wafer dicing where the wafer generates static electricity during the dicing saw cutting polishing process, and / or handling the wafer. It is recognized and understood that an automated / packaging automated device includes various post-wafer handling / packaging stages that can be a source of electrostatic discharge (ESD) to the conductors of the floating gate structure. In the absence of a connection to a silicon substrate (or semiconductor substrate) that provides an electrical path that drains such charge accumulation, the charge can be undesirable or damage to the gate oxide 165 (eg, charge injection into the oxide or underlying layer). Can build up to cause low level oxide breakdown to the substrate). Trap charges at the gate oxide or gate oxide-semiconductor interface can in turn cause undesirable and / or unpredictable changes in ISFET operation or performance.

  In view of the foregoing, other inventive aspects of the present disclosure are directed to methods and apparatus for improving ISFET performance by reducing trap charge or mitigating antenna effects. In one aspect, the trap charge may be reduced after manufacture of the sensor array, while in other aspects the manufacturing process itself is modified to reduce the charge that can be injected by some conventional process steps. May be. In still other embodiments, “in manufacturing” and “post manufacturing” techniques may be used in combination to reduce trap charge and thereby improve ISFET performance.

  With respect to manufacturing process changes to reduce trapped charge, in one aspect, the thickness of the gate oxide 165 shown in FIG. 11A may be specifically selected to drain charge stored on the substrate. In particular, thinner oxides can drain a sufficient amount of stored charge through the gate oxide to the substrate without being trapped. In another aspect based on this concept, the pixel may be designed to include an additional “sacrificial” device, ie, another transistor having a gate oxide thinner than the ISFET gate oxide 165. The ISFET floating gate structure may then be connected to the gate of the sacrificial device that functions as a “charge bleed-off transistor”. Of course, it should be understood that the price for including such a sacrificial device includes an increase in pixel size and complexity.

  In another aspect, the top metal layer 304 of the ISFET floating gate 170 shown in FIG. 11A is capped with a dielectric prior to plasma etching to mitigate trapped charges. As mentioned above, the charge stored in the floating gate structure can be combined in some cases by the plasma used for metal etching. Generally, a photoresist is applied over the etched metal and then patterned based on the desired shape for the underlying metal. In one exemplary implementation, a capping dielectric layer (eg, an oxide) is deposited on the metal being etched to provide an additional barrier to charge from the plasma etch process prior to photoresist application. It may be deposited. In one aspect, the capping derivative layer may remain on the back surface to form part of the passivation layer 172.

  In yet another aspect, the metal etch process of the top metal layer 304 may be modified to include wet chemistry or ion beam milling rather than plasma etching. For example, the metal layer 304 can be selectively etched into the underlying dielectric using water chemistry (see, eg, http://www.transene.com/aluminum.html, which is incorporated herein by reference). ). Another alternative approach is to use ion milling rather than plasma etching into the metal layer 304. Ion milling is commonly used for etching materials that cannot be easily removed by conventional plasma or wet chemistry. Since the ion milling process does not use an oscillating electric field like plasma, no charge increase occurs in the metal layer. Yet another metal etching alternative involves optimizing the plasma conditions to reduce the etch rate (ie, to a weaker power density).

  In yet another aspect, the metal layer may be restructured to facilitate complete electrical isolation during the definition of the floating gate. In one aspect, a large area ISFET floating gate is designed to stack metal so that nothing is connected during its final definition, an alternative metal layer as a “jumper” that senses the electrical connection of the transistor's floating gate. May be required. This “jumper” connection scheme prevents charge from flowing from the large floating gate to the transistor. This method may be implemented as follows (M = metal layer): i) connect M1 to the poly gate electrode, ii) connect M2 to M1, iii) M3 defines a floating gate and is isolated IV) island-like and connected remotely to M2, M4 “jumper” having a very small area etched on an isolated island and connected to the floating gate, M3 connecting the M3 floating gate to the polygate M1 / Immediately connect to M2 / M3 on the transistor active region, and v) the M3-M4 interlayer dielectric is removed only on the floating gate to expose the bare M3 floating gate. Among the methods briefly described above, ISFET structures according to some aspects described above leave M4 passivation in place on the M4 floating gate, so v) need not necessarily be performed. However, in one aspect, removal improves ISFET performance in other respects (ie sensitivity). In any case, the last chemical passivation layer may be a thin ion sensitive sputter deposited metal oxide layer. The jumper structure coating described above can be implemented in a standard CMOS manufacturing flow such that the first three metal layers (ie, M1, M2, and M3) are all used as floating gates.

  With respect to post-fabrication processes that reduce trap charge, in one aspect, “forming gas annealing” may be used as a post-fabrication process to mitigate the potential adverse effects of trap charge. In forming gas annealing, the CMOS ISFET device is heated in a hydrogen and nitrogen mixed gas. Hydrogen gas in the mixed gas diffuses into the gate oxide 165 and neutralizes certain trap charges. In one aspect, the forming gas anneal does not necessarily remove all gate oxide damage that can result from trapped charges. Rather, in some cases, partial neutralization of some trapped charge significantly improves ISFET performance. In a typical annealing process according to the present disclosure, the ISFET can be heated at about 400-425 degrees Celsius for about 30-60 minutes in a hydrogen / nitrogen gas mixture containing 10% -15% hydrogen. In one particular implementation, heating at 425 degrees Celsius for about 30 minutes in a hydrogen / nitrogen mixture containing 10% hydrogen was found to significantly improve ISFET performance. For aluminum CMOS processes, the annealing temperature should be kept at 450 degrees Celsius or less to avoid aluminum metallurgy damage. In another aspect of a typical annealing process according to the present disclosure, forming gas annealing reduces damage due to trap charges injected by the dicing process itself and / or other dicing pre-treatment stages (eg, metal plasma etching). This is done after the wafer on which the ISFET array is manufactured is diced.

  In yet another process according to aspects of the present disclosure to mitigate the potential negative effects of trapped charges, various “electrostatic discharge (ESD) -sensitive protocols” can be applied to various wafers. It may be applied at the handling / packaging stage of the post-manufacturing process. For example, in one exemplary process, an antistatic dicing tape may be used to secure the wafer substrate in place (eg, during the dicing process). Also, although high resistance (eg, 10 MΩ) deionized water is conventionally used to cool a dicing saw, one aspect of the present disclosure allows water with lower resistance / more conductivity to be charged through water. May be used for this purpose of promoting. For example, deionized water is treated with carbon dioxide to a lower resistance, improving the increased charge conduction from the dicing process. In addition, conductive and grounded die ejection jigs are used in various processes of wafer dicing / handling / packaging and to provide an effective conduction path for the charge generated during any of these stages. This reduces the chance of charge accumulation on one or more conductors of the floating gate structure of each ISFET in the array.

In yet another embodiment that includes a post-fabrication process that reduces trapped charge, the gate oxide region of the ISFET may be irradiated with ultraviolet light. Referring to FIG. 11A, in one exemplary implementation according to this aspect, an optional hole or window 302 is formed in the ISFET floating gate structure in the top metal layer 304 of each pixel of the array during ISFET array fabrication. It may be included in close proximity. If this window is formed, it is intended to allow UV light to enter the ISFET gate region. Particularly, various layers of the pixel 105 ₁ shown in FIGS. 11A and 12A~L an ultraviolet incident from the window 302, in a manner substantially unobstructed on the region adjacent to the polysilicon gate 164 and gate oxide 165 Configured to collide.

In order to facilitate the UV irradiation process to reduce trap charge, we generally consider that silicon nitride and silicon oxynitride absorb UV light significantly, so that other than silicon nitride and silicon oxynitride Is recognized and understood that it needs to be used in the passivation layer 172 shown in FIG. 11A. In view of the foregoing, this material needs to be replaced by another that is fairly transparent to ultraviolet light, examples include but are not limited to phosphosilicate glass (PSG) and borophosphosilicate glass (BPSG). . PSG and BPSG, however, pass hydrogen and hydrogen ions. Thus, for use in ISFET passivation layers designed for pH detection, PSG and BPSG do not pass ions such as aluminum oxide (Al ₂ O ₃ ) and are highly permeable to ultraviolet light. Together, they can be used to form a passivation layer. For example, referring again to FIG. 11A, PSG or BPSG can be substituted for silicon nitride or silicon oxynitride in the first portion of the passivation layer 172 and a thin layer of aluminum oxide (eg, 400-600 angstroms) is used. It may be used for the second portion of the passivation layer 172 (eg, aluminum oxide may be deposited using a post-CMOS lift-off lithography process).

から

の各信号は、同時に全ての行のセンサアレイを有効／選択（すなわち、ターンオン）し、これにより、アレイの全てのＩＳＦＥＴが各列における制御可能な電流ソース１０６_ｊと接続するように、生成される。各列の全てのピクセルが同時に選択されると、所与の列の電流ソース１０６_ｊは列の全てのピクセルによって共有される。列増幅器１０７Ａおよび１０７Ｂは、バイアス電源ＶＢ４を除去することによって無効となり、同時に、所与の列において各ＩＳＦＥＴのドレインと接続している増幅器１０７Ｂの出力は、制御信号「ＵＶ」に応答するスイッチを介して接地される。また、アレイの全てのＩＳＦＥＴのための列ボディ電圧Ｖ_ＢＯＤＹは、電気接地に接続される（すなわち、Ｖ_ＢＯＤＹ＝０）（上述のとおり、アレイの通常の動作中、ボディバイアス電圧Ｖ_ＢＯＤＹは、アレイに適用可能な最大電圧,例えば、ＶＤＤＡに接続することができる、）。１つの典型的な工程において、全ての制御可能な電流ソース１０６_ｊのためのバイアス電圧ＶＢ１は、各ピクセルのＩＳＦＥＴが約１μＡの電流で導電するように設定される。ＩＳＦＥＴアレイがこのようにバイアスされると、十分量の紫外線が照射される（例えば、約２０ミリワット／ｃｍ^２の照射を生成するＥＰＲＯＭイレーサーから、アレイから約１インチの距離で、約１時間）。照射後、アレイは、イオン濃度などの化学的特性の測定に用いる前に、数時間以上放置し、安定させてもよい。 In another aspect of the embodiment involving UV irradiation, each ISFET of the sensor array must be properly biased during UV irradiation to facilitate the reduction of trap charge. In particular, high energy photoelectrons from ultraviolet radiation impinging on the bulk silicon region 160 where the ISFET conductive channel is formed promotes neutralization of trapped charges in the gate oxide as current flows through the ISFET conductive channel. Create electron-hole pairs. For this purpose, the array controller shown below in FIG. 17 generates an appropriate signal to bias the array's ISFETs during the UV irradiation process. In particular, referring again to FIG.

From

Are simultaneously generated / enabled to enable / select (ie, turn on) all rows of sensor arrays, thereby connecting all ISFETs in the array to controllable current sources 106 _j in each column. The When all the pixels in each column are simultaneously selected, the current source 106 _j of a given column is shared by all pixels of the column. The column amplifiers 107A and 107B are disabled by removing the bias power supply VB4, and at the same time the output of the amplifier 107B connected to the drain of each ISFET in a given column has a switch responsive to the control signal “UV”. Is grounded. Also, the column body voltage V _BODY for all ISFETs in the array is connected to electrical ground (ie, V _BODY = 0) (as described above, during normal operation of the array, the body bias voltage V _BODY is The maximum voltage applicable to the array, eg, can be connected to VDDA). In one exemplary process, the bias voltage VB1 for all controllable current source 106 _j is, ISFET of each pixel is set to the conductive about 1μA of current. When the ISFET array is thus biased, a sufficient amount of UV radiation is applied (eg, from an EPROM eraser that produces about 20 milliwatts / cm ² of irradiation, at a distance of about 1 inch from the array for about 1 hour). . After irradiation, the array may be left to stabilize for several hours or more before being used to measure chemical properties such as ion concentration.

FIG. 13 illustrates a block diagram of an exemplary CMOS IC chip implementation of ISFET sensor array 100 based on the column and pixel design described above with respect to FIGS. 9-12, according to one inventive aspect of the present specification. In one aspect of this embodiment, the array 100 includes 512 columns from 102 ₁ to 102 ₅₁₂ with corresponding column bias / read circuits 110 ₁ to 110 ₅₁₂ (one in each column as shown in FIG. 9). ), Where each column includes ₅₁₂ geometrically square pixels from 105 ₁ to 105 ₅₁₂ , each having a size of approximately 9 micrometers × 9 micrometers (ie, the array is 512 columns × 512). Line). In another aspect, the entire array (including pixels with associated row and column selection circuitry and column bias / readout circuitry together) is a specific integrated circuit (ASIC) having dimensions of about 7 millimeters × 7 millimeters Can be manufactured on a semiconductor die. Although an array of 512 × 512 pixels is shown in the embodiment of FIG. 13, it will be understood that arrays according to other embodiments may be implemented with different numbers of rows and columns and different pixel sizes, as described below with respect to FIGS. Should be.

  Also, as noted above, arrays according to various aspects of the present invention may be fabricated according to conventional CMOS fabrication techniques, as well as modified CMOS fabrication techniques (eg, the various of the chemFET arrays shown herein). It is understood that the functional aspects may be made easier by further deposition of passivation materials, process steps to mitigate trapped charges, etc.) and other semiconductor manufacturing techniques beyond those used in conventional CMOS manufacturing. Should be. In addition, various lithographic techniques may be used as part of the array manufacturing process. For example, in one typical implementation, lithographic techniques may be used, where a properly designed block covers the edge of the step-and-repeat lithographic exposure on the wafer substrate by about 0.2 micrometers. "Stitched" as In a single exposure, the maximum size is usually about 21 mm × 21 mm. By selectively exposing different blocks (side, top and bottom, core, etc.), an oversized chip can be defined on the wafer (up to a maximum, in extreme cases, one chip on the wafer, generally “wafer” Called scale integration).

In one aspect of the array 100 shown in FIG. 13, the first and last two columns 102 ₁ , 102 ₂ , 102 ₅₁₁ and 102 ₅₁₂ , as well as the first two pixels 105 ₁ and 105 ₂ in each column 102 ₃ through 102 ₅₁₀ are used. And the last two pixels 105 ₅₁₁ and 105 ₅₁₂ (eg, two rows and two columns of pixels at the periphery of the array) can be configured as a “reference” or “dummy”. Referring to FIG. 11A, for the dummy pixels of the array, the metal layer 304 at the top of each dummy pixel's ISFET (finally coupled to the ISFET polysilicon gate 164) leads to the same metal layer of the other dummy pixels. It is made accessible as a chip terminal and can be connected to the reference voltage VREF. As described above with respect to FIG. 9, the reference voltage VREF can also be applied to the bias / read circuit of each column of the array. In some exemplary implementations described further below, the pre-test / evaluation data is obtained from the column based on application of a reference voltage VREF, selection and readout of dummy pixels, and / or direct application of VREF to each column buffer. Can be obtained based on readout (eg, via a CAL signal) to facilitate offset determination (eg, inter-pixel and inter-column variance) and array calibration.

In FIG. 13, various power supplies and bias power supplies required for array operation (as described above with respect to FIG. 9) are provided to the array via electrical connections (eg, pins, metal pads). Therefore, “power supply and bias connection” is expressed as a block 195. The array 100 of FIG. 13 also provides two parallel output signals Vout1 and Vout2 from the row selection shift register 192, two sets of column selection shift registers 194 _1,2 and an array showing sensor measurement results. an output driver 198 ₁ and 198 _2. Various power supply and bias voltages, control signals for the row and column select shift registers, and control signals for the column bias / read circuit shown in FIG. 13 are provided by the array controller, as described further below with respect to FIG. It also reads output signals Vout1 and Vout2 (and other optional status / diagnostic signals) from array 100. In another aspect of the array embodiment shown in FIG. 13, an array is configured such that multiple regions (eg, multiple columns) of the array can be read simultaneously via multiple parallel array outputs (eg, Vout1 and Vout2). This facilitates increasing the data acquisition rate, as further described below with respect to FIGS. Although FIG. 13 has two column select registers that acquire data from two columns simultaneously and parallel output signals Vout1 and Vout2, in other aspects, an array according to the present disclosure has only one measurement signal output, Alternatively, it may be configured to have three or more measurement signal outputs, and in particular, as described further below with respect to FIGS. 19-23, a dense array according to another aspect has four or five or more parallel measurement signal outputs. However, it is to be understood that different regions of the array may be configured to provide data via 4 or 5 or more outputs simultaneously.

から

として例示する）。 Figure 14 illustrates a row select shift register, FIG. 15 illustrates one 194 ₂ column select shift register and FIG. 16 illustrate one 198 ₂ output drivers of the array 100 shown in FIG. 13 According to one typical implementation. As shown in FIGS. 14 and 15, the row and column selection shift register is implemented as a series of D-type flip-flops connected to a digital circuit positive power supply VDDD and a digital power supply ground VSSD. In the row and column shift registers, the data signal is applied to the D input of the first flip-flop of each series, and the clock signal is simultaneously applied to the clock inputs of all flip-flops of that series. For each flip-flop, the “Q” output reproduces the D input at the clock signal transition (eg, falling edge). Referring to FIG. 14, the row selection shift register 192 includes 512 D-type flip-flops, where the first flip-flop 193 receives vertical data DV and all flip-flops receive the vertical clock signal CV. Receive. The Q output of the first flip-flop 193 provides the first row select signal RowSel ₁ and is connected to the D input of the next flip-flop in the series. As further described below with respect to FIG. 18, the Q output of the continuous flip-flop connects to the next D input in the sequence, and row select signals RowSel ₂ to RowSel ₅₁₂ are transitioned to the edge falling edge of the vertical clock signal CV. Provided by. The last row selection signal RowSel ₅₁₂ may be received as an optional output of the array 100 as an LSTV (Last STage Vertical) providing an indication (eg, for diagnostic purposes) that the last row of the array has been selected. Good. Although not specifically shown in FIG. 14, each row selection signal RowSel ₁ to RowSel ₅₁₂ is applied to a corresponding inverter, and the signals are used to validate a given pixel in each column (see FIG. 9).

From

As an example).

With respect to column select shift registers 194 ₁ and 194 ₂ , they are implemented in a manner similar to row select shift registers, each column select shift register including 256 series connected to a flip-flop, and an odd column or array of arrays. It is involved in making it possible to read from either of the even columns. For example, Figure 15 illustrates the column select shift register 194 _2, which column selection signals ColSel _2, ColSel _4, · · ·, via ColSel _512, can read from all of the even rows of successive array While another column select signal 194 ₁ is configured to allow reading from all odd columns of a continuous array (column select signals ColSel ₁ , ColSel ₃ ,... -ColSel ₅₁₁ ). Both column select signals are controlled by a horizontal data signal DH and a horizontal clock signal CH, as further described below with respect to FIG. 18, to provide respective column select signals. As shown in FIG. 15, the last column select signal ColSel ₅₁₂ is an optional LSTH (Last Stage Horizontal) providing an indication (eg, for diagnostic purposes) that the last row of the array has been selected. May be received as output.

Again referring to FIG. 7 for some time, we have implemented the design of Milgrew et al shown in FIG. 7 as described above with respect to FIGS. It recognizes and understands significant advances over the row and column decoders used in conventional designs including. In particular, for the row decoder 92 and column decoder 94 shown in FIG. 7, the complexity of implementing these components in an integrated circuit array design as the array size increases and further inputs to both decoders are required. Increase dramatically. For example, an array with 512 rows and columns as described above with respect to FIG. 13 requires 9 inputs (2 ⁹ = 512) per row and column decoder if such a scheme is used for row and column selection, as well In addition, an array having 7400 rows and 7400 columns requires 13 inputs (2 ¹³ = 8192) per row and column decoder, as described below with respect to other aspects. In contrast, the row and column select shift registers shown in FIGS. 14 and 15 do not require additional input signals with the array size, but rather require additional D-type flip-flops (this is a constant in CMOS processes). Implemented). Thus, the shift register implementation shown in FIGS. 14 and 15 provides an easily scalable solution for array row and column selection.

In the embodiment of FIG. 13, "odd" column selection shift register 194 ₁ provides odd column selection signal to the output driver 198 ₁ "odd", "even" column select shift register 194 _2, the even column selection signal " even "to provide to the output driver 198 _2. Both output drivers are configured similarly, an example of the even output driver 198 ₂ in FIG. 16. In particular, FIG. 16, each of the even column output signals _{V COL2, V COL4, ···,} V COL512 ( referred to as a column signal output _{V COLj} common names in FIG. 9), the corresponding switch ₁₉₁ 2, ₁₉₁ 4, _{., 191 512,} is applied to the even column select signal ColSel ₂ provided by the column select register 194 _2, ColSel _4, ..., in response to ColSel _512, the even column output signals and the buffer amplifier 199 (BUF ) Are continuously connected via the bus 175. In FIG. 16, a buffer amplifier 199 receives power from the output buffer positive power supply voltage VDDO and responds to an output buffer bias voltage VBO0 that controls a bias for buffer output. In view of the high impedance input of the buffer amplifier 199, a current sink 197 responsive to a bias voltage VB3 is the drive current suitable for the output of the column output buffer of the selected column (see the buffer amplifier 111 _j in FIG. 9) ( For example, it is connected to the bus 175 to provide about 100 μA order). Buffer amplifier 199 provides an output signal Vout2 based on the selected even column of the array, at the same time, referring to FIG. 13, the corresponding "odd" output driver 198 _1, based on the selected odd column of the array output A signal Vout1 is provided.

In one exemplary implementation, the even and odd switch output drivers 198 ₁ and 198 ₂ (e.g., shown in FIG. 16, a switch _{191 2, 191 4, ···,} 191 512), CMOS pair transmission gates ( 4), including an n-channel MOSFET and a p-channel MOSFET, and an inverter is connected to a given transmission gate switch 191 that allows each column select signal and its complementary signal to be switched. You may use so that it may be applied. Each switch 191 has a series resistance and, when enabled or “on”, connects the corresponding column output signal and bus 175, and similarly provides capacitance to bus 175 when the switch is OFF. Larger switches reduce series resistance, allow higher drive current for bus 175, and generally allow bus 175 to settle faster. On the other hand, a larger switch increases the capacitance of bus 175 and increases the stabilization time of bus 175 when the switch is off. Thus, there is a beneficial trade-off between switch series resistance and capacitance with respect to switch size.

The ability of bus 175 to quickly and stably enable subsequent sequential switches in sequence facilitates the acquisition of high speed data from the array. To this end, in some embodiments, the output drivers 198 ₁ and 198 ₂ of the switch 191 is specially configured to reduce the settling time of the bus 175 significantly. Both the n-channel and p-channel of a given switch increase the capacitance of the bus 175. However, n-channel MOSFETs generally have better frequency response and current drive capacity than p-channels. In view of the above, the inventors have sought to improve the stabilization time of bus 175 by implementing “asymmetric” switches with different sizes of n-channel and p-channel MOSFETs for a given switch. Recognize and understand that several excellent properties of n-channel MOSFETs are available.

For example, in one aspect, referring to FIG. 16, the current sink 197 is configured such that the bus 175 is normally when all switches 191 ₂ , 191 ₄ ,..., 191 ₅₁₂ are open or off (not conducting). It can be configured to be “pulled down (pulled down)”. Considering some limited expected signal dynamic range of the column output signal based on ISFET measurements, if a given switch is enabled or on (conducting), in many instances most of the conduction is switched This is performed by a CMOS pair of n-channel MOSFETs constituting Thus, in one aspect of this embodiment, the n-channel and p-channel MOSFETs of each switch 191 are sized differently. That is, in one typical implementation, the n-channel MOSFET is significantly larger than the p-channel MOSFET. More specifically, considering n-channel and p-channel MOSFETs of equal size as a reference point, in one implementation, n-channel MOSFETs are approximately 2-2.5 times larger in size and p-channel MOSFETs are reduced in size by about 8 to 10 times, and n-channel MOSFETs are about 20 times larger in size than p-channel MOSFETs. Due to the significant reduction in the size of the p-channel MOSFET and the increase in the size of the relatively small n-channel MOSFET, the overall capacitance of the off-state switch is significantly reduced, and the capacitance to the corresponding bus 175 is also significant. Decrease. At the same time, the larger n-channel MOSFET significantly improves current drive capability, frequency response and switch transconductance, and significantly reduces the stabilization time of bus 175.

The above examples, n- channel MOSFET is shows an asymmetric switch 191 for the output drivers 198 ₁ and 198 ₂ greater than p- channel MOSFET, in another aspect, the reverse, i.e., p- channel MOSFET is n- It should be understood that an asymmetric switch larger than the channel MOSFET may be implemented. In another aspect of this embodiment, referring again to FIG. 16, current sink 197, as a current source for properly driving the output of the column output buffer of the selected column (see the buffer amplifier 111 _j in FIG. 9) The bus 175 is normally “pulled down”, which may function alternatively, and when all switches 191 ₂ , 191 ₄ ,..., 191 ₅₁₂ are open or off (not conducting). May be configured. In this situation, most switch conduction can be achieved by the CMOS pair of p-channel MOSFETs that make up the switch. Because the frequency response of the p-channel MOSFET is lower than that of the n-channel MOSFET, the overall beneficial effect of reducing the stabilization time of the bus 175 is somewhat less than that described above, but the switch capacitance is reduced (and this). The advantage of reducing the bus capacity due to the above can be understood by this aspect. However, the asymmetric switch based on a larger p-channel MOSFET still significantly facilitates the reduction of bus stabilization time, and the column output buffer amplifier (111j in FIG. 9) with a significantly increased gain, “body junction” -tied) "can also provide a circuit implementation that can be a source follower.

In yet another aspect relating to promoting fast stabilization of bus 175 shown in FIG. 16, it should be understood that fewer switches connected to bus 175 result in smaller bus capacitance. With this in mind, referring again to FIG. 13, in yet another embodiment, three or more output drivers 198 ₁ and 198 _2, as each output driver handles a number of columns less arrays, ISFET array 100 You may use for. For example, rather than having all even columns processed by one driver and all odd columns processed by one driver, each output driver is a quarter or even a half of all columns of the array. The array may have four column select registers 194 ₁₂₃₄ and corresponding four output drivers 198 ₁₂₃₄ . In such an implementation, each output driver will therefore halve the number of switches 191 and each output driver's bus 175 will have a corresponding lower capacitance compared to the embodiment described above with respect to FIG. Thus, the bus stabilization time is improved. In this example, four output drivers are shown for illustrative purposes, but the present disclosure is not limited in this respect, and in the situation described above, there are virtually any number of output drivers far beyond two. It should be understood that it may be used to improve bus stabilization time. Other array aspects in which more than two output drivers are used to facilitate high speed data acquisition from the array are shown in more detail below (eg, with respect to FIGS. 19-23).

  In one aspect of the array design described above with respect to FIGS. 13-16, remote analog power supply voltage connections (between VDDA and VSSA), digital power supply voltage connections (between VDDD and VSSD), and output buffer power supply voltage connections (between VDDO and VSSO). Are provided on the array to facilitate noise blocking and reduced crosstalk between the various array components, which allows the signal-to-noise ratio (SNR) of the Vout1 and Vout2 output signals to be reduced. Be improved. In one typical implementation, the positive power supply voltages VDDA, VDDD, and VDDO each may be about 3.3 volts. In another aspect, these voltages are each provided “off-chip” by one or more programmable voltage sources, as described further below with respect to FIG.

  FIG. 17 illustrates a block diagram of the sensor array 100 of FIG. 13 connected to the array controller 250 in accordance with one inventive aspect of the present disclosure. In various exemplary aspects, as described above with respect to FIG. 8, the array controller 250 may be manufactured as a “stand alone” controller or compatible “card” that forms part of the computer 260. In one aspect, as shown in FIG. 17, the functionality of the array controller 250 can be controlled by the computer 260 via the interface block 252 (eg, serial interface via USB port or PCI bus, Internet connection, etc.). . In one aspect, the array controller 250 is manufactured as a printed circuit board to which the array 100 is connected, similar to a conventional IC chip (eg, the array 100 is configured as an ASIC that is connected to the array controller). One aspect of such an embodiment may be implemented as a field programmable gate array (FPGA) in which all portions of the array controller are configured to perform the various array controller functions described in more detail below.

  In general, the array controller 250 provides various power supply and bias voltages to the array 100 as well as signals related to row and column selection, pixel output sampling and data acquisition. In particular, the array controller 250 reads two analog output signals Vout1 (odd columns) and Vout2 (even columns) containing each combined pixel voltage signal from the array 100 and provides each pixel with measurement data to provide to the computer 260. Digitize the signal. In some aspects, the array controller 250 may also be configured to perform or facilitate array calibration and functional diagnostics, and any UV irradiation process, as described above with respect to FIG. 11A.

As illustrated in FIG. 17, the array controller 250 generally connects the array 100 to an analog power supply voltage and ground (VDDA, VSSA), a digital power supply voltage and ground (VDDD, VSSD), and a buffer power supply voltage and ground (VDDO, VSSO). )I will provide a. In one typical implementation, the positive supply voltages VDDA, VDDD and VDDO are each about 3.3 volts. As described above, in one aspect, each of these power supply voltages is provided to the array 100 via separate conductive paths to facilitate noise rejection. In another aspect, these power supply voltages are generated from respective power supplies / regulators, or one or more of these power supply voltages are generated from a common common source within power supply 258 of array controller 250. Also good. The power supply 258 may also provide bias voltages (eg, VB1, VB2, VB3, VB4, VBO0, V _BODY ) required for array operation and a reference voltage VREF used for array diagnostics and calibration. In another aspect, the power supply 258 can be controlled by the computer 260 to allow some or all of the bias voltage, reference voltage, and power supply voltage to be changed under software control (ie, programmable bias settings) 1 Or two or more digital-to-analog converters (DACs). For example, power supply 258 in response to computer control includes bias voltages VB1 and VB2 for pixel drain current, VB3 for column bias drive, VB4 for column amplifier bandwidth, and VBO0 for column output buffer current drive. Adjustment can be facilitated. In some aspects, one or more bias voltages can be adjusted to optimize the settling time of a signal from an effective pixel. In addition, the body voltage V _BODY common to all ISFETs in the array is grounded during post-production UV processing to reduce trap charge and then during normal operation for diagnostic analysis, calibration and measurement / data collection. , Can be connected to a high voltage (eg, VDDA). Similarly, the reference voltage VREF can be changed to facilitate a variety of diagnostics and calibrations.

  As also shown in FIG. 17, a reference electrode 76 that is typically used in connection with the analytical solution measured by the array 100 (as described above in connection with FIG. 1) provides a reference potential for the pixel output voltage. A power source 258 may be connected. For example, in one implementation, the reference electrode 76 may be connected to the power supply ground (eg, analog ground VSSA) to provide a pixel output voltage reference based on equation (3) above. In other exemplary implementations, the reference electrode voltage is determined by placing a solution / sample of known pH level in proximity to the sensor array 100, and the reference electrode voltage is desired by the array output signals Vout1 and Vout2. By adjusting until the reference level is reached, the subsequent change in pixel voltage reflects a local change in pH relative to a known reference pH level. In general, in some aspects, the reference voltage VREF provided by the power supply 258 is used to set the reference electrode 76, but the voltage associated with the reference electrode 76 is the reference voltage VREF described above (various array diagnostics and It should be understood that it is not necessarily the same as that used for the calibration function.

  With respect to data acquisition from the array 100, in one aspect, the array controller 250 of FIG. 17 further collects the output signals Vout1 and Vout2 from the sensor array into a buffer and provides one or more preamplifiers that provide selectable gain. 253 may be included. In one aspect, the array controller 250 may include one preamplifier for each output signal (eg, two preamplifiers for two analog output signals). In other aspects, the preamplifier is configured to receive an input voltage between 0.0 and 3.3 volts, and a programmable / computer selectable gain (eg, 1, 2, 5, 10 and 20) and It may have a low noise output (eg, <10 nV / sqrtHz) and provide a reduced pass filter (eg, 5 MHz and 25 MHz bandwidth). In yet another aspect, the preamplifier has a programmable / computer-selectable offset to the input and / or output voltage signal to set a defined level for the desired range.

  The array controller 250 of FIG. 17 also provides one or more analog-to-convert sensor array output signals Vout1 and Vout2 to digital outputs (eg, 10 or 12 bits) to provide data to the computer 260. A digital converter (ADC) 254 may be included. In one aspect, one ADC may be used for each analog output of the sensor array, and each ADC may be connected to a corresponding preamplifier (if a preamplifier is used in a given implementation). ). In another aspect, the ADC has a computer selectable input range (eg, 50 mV, 200 mV, 500 mV, 1 V) to facilitate compatibility with different ranges of array output signals and / or preamplifier parameters. May be. In yet another aspect, the ADC bandwidth may be greater than 60 MHz and the data acquisition / conversion rate may be greater than 25 MHz (eg, 100 MHz or higher).

  In the embodiment of FIG. 17, ADC acquisition timing and array row and column selection are controlled by timing generator 256. In particular, the timing generator provides digital vertical data and clock signals (DV, CV) to control row selection and digital horizontal data and clock signals (DH, CH) to control column selection. The column sampling and holding signal COL SH provides the pixel voltage of each sample to the valid row, as described above in connection with FIG. In some aspects, the timing generator 256 may be implemented by a microprocessor that executes code and is configured as a multi-channel digital pattern generator to form an appropriately time-controlled signal. In one exemplary implementation, the timing generator 256 may be implemented as a field programmable gate array (FPGA).

  FIG. 18 illustrates an exemplary timing diagram for such a signal provided by timing generator 256 to obtain pixel data from sensor array 100. For purposes of the following description, “frame” is defined as the data set that contains a value for each pixel in the array, and “frame rate” is defined as the rate at which successive frames can be obtained from the array. In the example of FIG. 18, a typical frame rate of 20 frames / second is selected to illustrate the operation of the array (ie, row and column selection and signal acquisition). However, arrays and array controllers according to the present disclosure are not limited in this regard and may have lower frame rates (eg, 1-10 frames / second) or higher frame rates (eg, 25, 30, 40, 50, 60, 70). It should be understood that different frame rates, including ˜100 frame rates / second, etc.), equivalent or higher pixel counts are possible. In some implementations, a data set may be acquired that includes many frames over a few seconds conducted to a given analyte or experiment of analyte (s). Some such experiments are performed in series, possibly with data transfer / processing and / or washing of the sensor array ASIC and a pause to prepare reagents for subsequent experiments.

を介して有効にされる。有効なピクセルは、ある期間一定化され、ＣＯＬ ＳＨ信号は、各列のサンプリング／保持スイッチを閉じ、また列の第１のピクセルによる出力電圧値を列のサンプリング／保持キャパシタＣ_ｓｈ上に保持するために、一時的に（briefly）アサートされる。この電圧はその後列出力電圧Ｖ_ＣＯＬｊとして２つ（奇数および偶数列）のアレイ出力ドライバ１９８_１および１９８_２のうち１つに印加される（例えば、図１６参照）。ＣＯＬ ＳＨ信号は、その後ディアサートされ、これにより、各列のサンプリング／保持スイッチが開き、また列増幅器１０７Ａおよび１０７Ｂから、列出力バッファ１１１_ｊが分断（decoupling）される。その後すぐに、ピクセルの第２の行が行選択信号

を介して有効にされる。ピクセルの第２の行が安定化されている間、行選択信号は、第１の行に関連する列出力電圧を読むために、２つずつ生成される（１つの奇数および１つの偶数、奇数列選択信号は連続して偶数出力ドライバに入力され、偶数行列選択信号は連続して奇数出力ドライバに入力される）。こうして、アレイ内の所与の行が有効にされ安定化されている間に、前の行が２列づつ読み出される。行選択およびサンプリング／読取りを交互に配置することおよび所与の行に対して複数列ずつ読むことによって、データフレームは、顕著に合理化された方法でアレイから取得することができる。 In one implementation, the array controller 250 controls the array 100 with rows enabled one at a time. For example, referring again to FIG. 9 for a while, the first row of pixels is the row selection signal.

Enabled through. The valid pixels are stabilized for a period of time, and the COL SH signal closes the sampling / holding switch for each column and holds the output voltage value from the first pixel in the column on the column sampling / holding capacitor C _sh. In order to be asserted temporarily (briefly). This voltage is applied to one of the array output drivers 198 ₁ and 198 ₂ of the two as the rear output voltage _{V COLj} (odd and even columns) (e.g., see FIG. 16). The COL SH signal is then deasserted, thereby opening the sampling / holding switch for each column and decoupling the column output buffer 111 _j from the column amplifiers 107A and 107B. Immediately thereafter, the second row of pixels becomes

Enabled through. While the second row of pixels is stabilized, row selection signals are generated in duplicate to read the column output voltage associated with the first row (one odd and one even, odd The column selection signal is continuously input to the even output driver, and the even matrix selection signal is continuously input to the odd output driver). Thus, while a given row in the array is validated and stabilized, the previous row is read by two columns. By alternating row selection and sampling / reading and reading multiple columns for a given row, data frames can be obtained from the array in a significantly streamlined manner.

  FIG. 18 illustrates a timing diagram of the process for a typical frame rate of 20 frames / second. Considering this frame rate and 512 rows in the array, each row must be read out in approximately 98 microseconds, as shown by the vertical depiction in FIG. Thus, the vertical clock signal CV has a period of 98 microseconds (i.e., a clock frequency greater than 10 kHz), with the activation of a new row by the trailing edge (negative conduction) of the CV signal. The left side of FIG. 18 shows the beginning of a new frame, where the vertical data signal DV is asserted before the first trailing edge of the CV signal and deasserted before the trailing edge of the next CV signal (continuous). The vertical data signal is reasserted again only after row 512 has been enabled for data acquisition from the frame to be). Also, immediately before the trailing edge of each CV signal (eg, a new row is enabled), the COL SH signal is asserted for 2 microseconds, leaving 50 nanoseconds before the trailing edge of the CV signal.

In FIG. 18, the first generation of the COL SH signal is actually sampling the pixel values in row 512 of the array. Thus, when the first trailing edge of the CV signal occurs, the first row is enabled and stabilized until the second generation of the COL SH signal (approximately 96 microseconds). During the first row stabilization time, the pixel values in row 512 are read via a column select signal. Since two column selection signals are generated simultaneously to read 512 columns, the horizontal clock signal CH must be generated 256 cycles within this period, and each trailing edge of the CH signal has one odd and One even column selection signal is generated. As shown in FIG. 18, the first trailing edge of the CH signal in a given row occurs after 2 microseconds of row selection to stabilize the voltage value held on the sampling / holding capacitor C _sh. Time-controlled and provided by a column output buffer. Also for each row, the horizontal data signal DH is asserted before the first trailing edge of the CH signal and deasserted before the next trailing edge of the CH signal. The last two columns (eg, 511 and 512) are selected before the occurrence of the COL SH signal, which occurs approximately 2 microseconds before the next row becomes valid, as described above. Thus, 512 columns are read in duplicate within about 94 microseconds (ie 98 microseconds / row, minus the first and last 2 seconds of each row). This results in a data rate of about 2.7 MHz for each array output signal Vout1 and Vout2. In another aspect, the ADC (s) 254 is controlled by the timing generator 256 to sample the output signals Vout1 and Vout2 at a significantly higher rate to provide a plurality of digitized samples for each pixel measurement. May be controlled and then averaged (e.g., the ADC data acquisition rate can be about 100 MHz taking a 2.7 MHz array output signal, so that about 35-40 samples per pixel measurement. Can be provided.)

In addition to controlling the sensor array and ADC, the timing generator 256 may be configured to facilitate various array calibrations and functional diagnostics as well as optional ultraviolet irradiation processes. For this purpose, the timing generator can utilize a signal LSTV indicating the selection of the last row of the array and a signal LSTH indicating the selection of the last column of the array. The timing generator 256 generates a CAL signal that inputs the reference voltage VREF to the column buffer amplifier, and generates a UV signal that grounds the drains of all ISFETs in the array during the UV irradiation process (see FIG. 9). Can also be involved. The timing generator can also provide a control function on the power supply 258 to properly control the power supply or bias voltage during various calibrations and functional diagnostics or UV irradiation. For example, during UV irradiation, the timing generator can control the power supply for connecting the body voltage V _BODY and ground while the UV signal is activated to ground the ISFET drain. With respect to array calibration and diagnostics, as well as UV illumination, in some implementations, the timing generator can receive specialized programs from a computer 260 that provides appropriate control signals. In one aspect, the computer 260 can determine various data obtained from the dummy pixels of the array, as well as column information based on the application of the CAL signal and the reference voltage VREF, and determine various calibration parameters for a given array and / or. Or it can be used to create specialized programs for calibration and functional diagnosis.

  With respect to the computer interface 252 of the array controller 250, in one exemplary implementation, the interface is configured to facilitate a data rate of approximately 200 MB / sec to the computer 260. The computer 260 is configured to receive data at 200 MB / sec and process the data to reconstruct the pixel image (eg, shown in pseudo color on the monitor). For example, a computer may be configured to execute a general purpose program including routines written in C ++ or Visual Basic for data manipulation, and a display is desired.

  Having described some aspects of a typical ISFET array and array controller according to the present disclosure, FIGS. 19-23 are alternative CMOS of ISFET sensor arrays with higher pixel counts according to yet another inventive aspect. 1 illustrates a block diagram of an IC chip. In one aspect, each ISFET array shown further below with respect to FIGS. 19-23 may be controlled by a controller similar to that shown in FIG. 17, and in order to accommodate a larger number of pixels, In some cases, minor changes may be involved (eg, additional preamplifier 253 and analog to digital converter 254).

FIG. 19 shows a block diagram of an ISFET sensor array 100A based on the column and pixel design described above in connection with FIGS. 9-12 and a 0.35 μm COMS manufacturing method, according to one aspect of the present invention. Array 100A includes 2048 columns 102 ₁ -102 ₂₀₄₈ , where each column includes 2048 geometric square pixels 105 ₁ -105 ₂₀₄₈ , each having dimensions of approximately 9 μm × 9 μm. . Thus, the array includes over 4 million pixels (> 4 megapixels), and in one exemplary implementation, the complete array (ISFET pixels and associated circuitry) is approximately 20.5 mm by 20.5 mm in size. Can be manufactured as an integrated circuit chip.

In one aspect of the embodiment shown in FIG. 19, the array 100A may be configured, at least in part, as a plurality of groups of individually controllable pixels. For example, each column of pixels can be divided into a top portion and a bottom portion, and the accumulation of pixels at each top portion of the column forms a _first group of rows 400 ₁ (eg, top group, rows: 1-1024). , The accumulation of pixels at each bottom portion of the column forms a _second group 400 _{2 of} rows (eg, bottom group, rows: 1025-2048). Next, each of the first and second groups of rows (eg, top and bottom) is associated with a corresponding row select register, column bias / read circuit, column select register, and output driver. In this manner, pixel selection and data acquisition from the first group 400 ₁ and the second of each of the group 400 ₂ line, a pixel selection and data acquisition from the entire array 100 shown in FIG. 13, substantially similar It becomes. In other words, in one aspect, the array 100A of FIG. 19 includes a “subarray” of two different groups of pixels that are controlled simultaneously, thereby providing efficient acquisition of data from multiple pixels.

から

までの信号を生成し、第２の行選択レジスタ１９２_２は、

から

までの信号を生成することができる。他の態様において、両方の行選択レジスタ１９２_１および１９２_２は、共通の垂直クロックおよびデータ信号を同時に受信してもよく、こうして、アレイの2つの行が所与の時間において、トップ群から１つ、ボトム群からもう１つを有効化される。 In particular, FIG. 19, the first row selection group 400 ₁ of row is controlled by a first row select register 192 _1, the second row selection of the group 400 ₂ rows by ₂ second row select register 192 Indicates that it may be controlled. In one aspect, each of the row select registers 192 ₁ and 192 ₂ may be configured as described above in connection with FIG. 14, receiving a vertical clock (CV) signal and a vertical data (DV) signal, generating a row selection signal in response to; for example, a first row select register 192 _1,

From

Generates a signal to the second row select register 192 _2,

From

Signals up to can be generated. In other aspects, both row select registers 192 ₁ and 192 ₂ may receive a common vertical clock and data signal at the same time, so that two rows of the array are one from the top group at a given time. One is activated from the bottom group.

For each of the first and second groups of rows, the array 100A of FIG. 19 further includes column bias / read circuits 110 _{1T -110 2048T} (for the first row group 400 ₁ ) and 110 _1B -110. comprises _{2048 B} (for the second row group 400 _2), thus each column includes two instances of the bias / readout circuitry 110j shown in FIG. Array 100A also for the second row group 400 _2, comprises two column select registers _{192 1, 2} (odd and even) and two output drivers _{198 1, 2} (odd and even), also, the first to the row group 400 _1, two column select registers _{192 3,4} (odd and even) and two output drivers _{198 3,4} (odd and even) (i.e., four column select registers and a total of 4 Output drivers). The column select register has horizontal clock signals (CHT and CHB for the first row group and second row group, respectively) and horizontal data signals (for the first row group and second row group, respectively). DHT and DHB) are received to control the selection of odd and even columns. In one implementation, the CHT and CHB signals may be provided as a common signal, and the DHT and DHB may be provided as a common signal, thus reading four columns at a time from the array simultaneously (ie, one from each row group). Two odd columns and one even column). In particular, as described above in connection with FIGS. 13-18, two columns are read simultaneously for each valid row and the corresponding pixel voltages are provided as two output signals. Thus, through validating two rows at any given time and reading two columns for one row at any given time, the array 100A can generate four simultaneous output signals Vout1, Vout2, Vout3 and Vout4 can be supplied.

In one exemplary implementation of the array 100A of FIG. 19, wherein a complete data frame 20 frames / sec frame rate (first row group 400 ₁ and the second all pixels from both the line group 400 ₂₎ Acquire and validate 1024 pairs of rows in sequence for about 49 microseconds each. For each valid row, 1024 pixels are read by each column select register / output driver for approximately 45 microseconds (allowing 2 microseconds at the beginning and end of each row as described above in connection with FIG. 18). Thus, in this example, each of the array output signals Vout1, Vout2, Vout3 and Vout4 has a data rate of about 23 MHz. Again, it should be understood that in other implementations, data may be acquired from the array 100A of FIG. 19 at a frame rate other than 20 frames / second (eg, 50-100 frames / second).

  Like yet the array 100 of FIG. 13, in yet another aspect, the array 100A of FIG. 19 may include multiple rows and multiple columns of dummy or reference pixels 103 around the array to provide preliminary test / evaluation data, offset determination And / or facilitate array calibration. In addition, various power supplies and bias voltages (as described above in connection with FIG. 9) required for array operation are applied to array 100A in block 195 in a manner similar to that described above in connection with FIG. To supply.

  FIG. 20 shows a block diagram of an ISFET sensor array 100B based on a 0.35 μm COMS fabrication method and having a structure substantially similar to the array 100A described above in FIG. 19 according to yet another aspect of the present invention. Array 100B is also generally based on the column and pixel design described above in connection with FIGS. 9-12, but the dimensions / pitch of the pixels in array 100B are smaller than that of the pixels shown in FIG. With particular reference to FIGS. 10 and 11 again, the dimension “e” shown in FIG. 10 is about 9 μm in the embodiment of FIG. 20 without affecting the integrity of the active pixel elements located in the central region of the pixel. It is greatly reduced to about 5 μm. Similarly, the dimension “f” shown in FIG. 10 is reduced from about 7 μm to about 4 μm. In other words, some of the surrounding area of the pixel surrounding the active element impairs the layout and design of the top and cross-sections of the active element of the pixel shown in FIGS. But substantially (significantly) reduced. A top view of such a pixel 105A is shown in FIG. 20A, where dimension “e” is 5.1 μm and dimension “f” is 4.1 μm. In one aspect of this pixel design, pixel 105A includes fewer body connections B (eg, each corner of the pixel) compared to the pixel shown in FIG. 10 to facilitate dimensional reduction. This includes several body connections B all around the pixel.

As noted in FIG. 20, array 100B includes 1348 columns 102 ₁ -102 ₁₃₄₈ , where each column includes 1152 geometrically square pixels 105A ₁ -105A ₁₁₅₂ , each of which is approximately It has dimensions of 5 μm × 5 μm. Thus, the array includes more than 1.5 million pixels (> 1.5 megapixels), and in one exemplary implementation, the complete array (ISFET pixels and associated circuitry) has a dimension of about 9 mm × 9 mm. It can be manufactured as an integrated circuit chip. Like the array 100A of FIG. 19, in one aspect, the array 100B of FIG. 20, as described above in connection with FIG. 19, it is divided into groups 400 ₁ and 400 ₂ of the two rows. In one exemplary implementation, complete data frames (all pixels from the first row group 400 ₁ and the second row group 400 ₂ both) acquired at a frame rate of 50 frames / sec, thereby 576 pairs It is necessary to validate the rows in sequence for about 35 microseconds each. For each valid row, 674 pixels are read by each column select register / output driver for approximately 31 microseconds (allowing 2 microseconds at the beginning and end of each row as described above in connection with FIG. 18). Thus, in this example, each of the array output signals Vout1, Vout2, Vout3 and Vout4 has a data rate of approximately 22 MHz. Again, it should be understood that in other implementations, data may be obtained from the array 100B of FIG. 20 at a frame rate other than 50 frames / second.

FIG. 21 is an ISFET sensor array based on a 0.35 μm COMS fabrication method, incorporating the smaller pixel dimensions described above in connection with FIGS. 20 and 20A (5.1 μm square pixels), according to yet another aspect of the present invention. A block diagram of 100C is shown. As mentioned in FIG. 21, array 100C includes 4000 columns 102 ₁ -102 ₄₀₀₀ , where each column includes 3600 geometric square pixels 105A ₁ -105A ₃₆₀₀ , each of which It has a dimension of about 5 μm × 5 μm. The array thus includes more than 14 million pixels (> 14 megapixels), and in one exemplary implementation, the complete array (ISFET pixels and associated circuitry) is an integrated circuit chip having dimensions of approximately 22 mm × 22 mm. Can be manufactured as. Like the arrays 100A and array 100B of FIG. 19 and FIG. 20, in one aspect, the array 100C in FIG. 21 is divided into groups 400 ₁ and 400 ₂ of the two rows. However, unlike arrays 100A and 100B, for each group of rows, array 100C includes 16 column select registers and 16 output drivers, which simultaneously read 16 pixels at a time in a valid row, thus 32 Output signals Vout1 to Vout32 can be supplied from the array 100C. In one exemplary implementation, complete data frames (first row group 400 ₁ and the second all pixels from both the line group 400 ₂₎ can be acquired at a frame rate of 50 frames / sec, thereby 1800 pairs It will be necessary to validate the rows in sequence for approximately 11 microseconds each. For each valid row, 250 pixels (4000/16) are read by each column select register / output driver for approximately 7 microseconds (allowing 2 microseconds at the beginning and end of each row). Thus, in this example, each of the array output signals Vout1-Vout32 has a data rate of about 35 MHz. As in the previous aspect, it should be understood that in other implementations, data may be obtained from array 100C at a frame rate other than 50 frames / second.

  Although the exemplary array described above in connection with FIGS. 13-21 is based on a conventional COMS manufacturing method of 0.35 μm, the array according to the present disclosure is not limited in this respect because the shape is less than 0.35 μm This is because such an array may be manufactured using a CMOS manufacturing method (for example, 0.18 μm COMS processing technology). Thus, an ISFET sensor array with a pixel size / pitch significantly smaller than 5 μm can be manufactured, providing a very high density ISFET array. For example, FIGS. 22 and 23 are block diagrams of ISFET sensor arrays 100D and 100E, respectively, according to yet another aspect, based on a 0.18 μm COMS manufacturing method, where a pixel size of 2.6 μm is realized. The pixel design itself is substantially based on the pixel 105A shown in FIG. 20A, but at a smaller scale due to the 0.18 μm COMS method.

The array 100D of FIG. 22 includes ₂₈₀₀ columns 102 _{1 to} 102 ₂₈₀₀ , where each column is 2400, geometrically square, each having dimensions of about 2.6 μm × 2.6 μm. Including. Thus, the array includes more than 6.5 million pixels (> 6.5 megapixels), and in one exemplary implementation, the complete array (ISFET pixels and associated circuitry) is an integrated having dimensions of about 9 mm × 9 mm. It can be manufactured as a circuit chip. Like the array 100A, 100B and 100C in FIG. 19 to 21, in one aspect, an array 100D of FIG. 22 is divided into groups 400 ₁ and 400 ₂ of the two rows. However, unlike arrays 100A, 100B, and 100C, for each group of rows, array 100D includes 8 column select registers and 8 output drivers to simultaneously read 8 pixels at one time in a valid row, thus 16 Output signals Vout1 to Vout16 can be supplied from the array 100D. In one exemplary implementation, complete data frames (first row group and second row group 400 ₁ and 400 all pixels from ₂ both) can be acquired at a frame rate of 50 frames / sec, thereby 1200 pairs In turn, each for approximately 16-17 microseconds. For each valid row, 350 pixels (2800/8) are read by each column select register / output driver for approximately 14 microseconds (allowing 1-2 microseconds at the beginning and end of each row). Thus, in this example, each of the array output signals Vout1-Vout16 has a data rate of approximately 25 MHz. As in the previous aspect, it should be understood that in other implementations, data may be obtained from array 100D at a frame rate other than 50 frames / second.

The array 100E of FIG. 23 includes ₇₄₀₀ columns 102 _{1 to} 102 ₇₄₀₀ , where each column is 7400 geometrically square, each having a dimension of about 2.6 μm × 2.6 μm. Including. Thus, the array includes more than 54 million pixels (> 54 megapixels), and in one exemplary implementation, the complete array (ISFET pixels and associated circuitry) is an integrated circuit chip having dimensions of approximately 21 mm × 21 mm. Can be manufactured as. Like the array 100A~100D of 19-22, in one aspect, the array 100E of FIG. 23 is divided into groups 400 ₁ and 400 ₂ of the two rows. However, unlike arrays 100A-100D, for each row group, array 100E includes 32 column select registers and 32 output drivers, which simultaneously read 32 pixels at a time in a valid row, thus providing 64 Output signals Vout1 to Vout64 can be supplied from the array 100E. In one exemplary implementation, to get a complete data frame 100 frames / sec frame rate (first row group 400 ₁ and the second all pixels from both the line group 400 _2), thereby 3700 pairs It will be necessary to validate the rows in sequence for approximately 3 microseconds each. For each valid row, 230 pixels (7400/32) are read by each column select register / output driver for approximately 700 nanoseconds. Thus, in this example, each of the array output signals Vout1-Vout64 has a data rate of about 328 MHz. As in the previous aspect, it should be understood that in other implementations, data may be obtained from array 100E at a frame rate other than 100 frames / second.

Thus, in various examples of ISFET arrays based on the inventive concepts disclosed herein, an array pitch of about 9 μm (eg, less than 10 μm × 10 μm in sensor surface area) is 256,000 pixels (ie, Can be manufactured on a 7 mm x 7 mm semiconductor die with associated row and column selection and bias / read electronics, and more than 4 million pixels A similar sensor array (ie, 2048 × 2048 array, greater than 4 megapixels) can be fabricated on a 21 mm × 21 mm die. In other embodiments, an array pitch of about 5 μm allows an ISFET array and related electronic devices comprising about 1.55 megapixels (ie, 1348 × 1152 array) to be fabricated on a 9 mm × 9 mm die. It also allows ISFET arrays and associated electronic devices containing more than 14 megapixels to be fabricated on a 22 mm × 22 mm die. In yet another implementation, an ISFET sensor array having a pixel size / pitch significantly smaller than 5 μm can be manufactured using a CMOS manufacturing method that allows dimensions of less than 0.35 μm (eg, 0.18 μm CMOS processing technology). Can (eg, 2.6 μm array pitch, or pixel / sensor area less than 8 or 9 μm ² ), providing a significantly higher density ISFET array.

  In the ISFET array embodiment described above, the array pixels use p-channel ISFETs as described above in connection with FIG. However, ISFET arrays according to the present disclosure are not limited in this regard, and other aspects of pixel design for ISFET arrays may be based on n-channel ISFETs. In particular, any of the arrays described above in connection with FIGS. 13 and 19-23 can be implemented using pixels based on n-channel ISFETs.

For example, FIG. 24 shows the pixel design of FIG. 9 implemented using an n-channel ISFET and an accompanying n-channel MOSFET in accordance with another inventive aspect of the present disclosure. More specifically, FIG. 24 shows one exemplary pixel 105 ₁ (ie, the first pixel in the column) of the array column, along with column bias / read circuit 110j, where ISFET 150 (Q1) is n− Channel ISFET. Similar to the pixel design of FIG. 9, the pixel design of FIG. 24 includes only three elements: two n n responsive to ISFET 150 and one of the n row select signals (RowSel _{1 to} RowSel _n , positive logic). -Channel MOSFET switches Q2 and Q3. In the pixel of FIG. 24, no transmission gate is required and all devices in the pixel are “same type”, ie, n-channel devices. Further, similar to the pixel design of FIG. 9, to operate the three elements of pixel 105 ₁ shown in FIG. 24, only four signal lines per pixel, ie, lines 112 ₁ , 114 ₁ , 116 ₁ and 118 ₁ is necessary. In other respects, the pixel designs of FIGS. 9 and 24 both have a constant drain current I _Dj and a constant drain-source voltage V _DSj to obtain the output signal V _Sj from the effective pixel. They are similar in that they are structured.

One of the main differences between the n-channel ISFET pixel design of FIG. 24 and the p-channel ISFET design of FIG. 9 is that the current through the pixel is in the reverse direction. Thus, in FIG. 24, element 106 _j is a controllable current sink coupled to analog circuit voltage ground VSSA, and element 108 _j of bias / read circuit 110 _j is a controllable current source coupled to analog positive supply voltage VDDA. It is. Further, the body connection of ISFET 150 is not coupled to its source, but instead is coupled to the body connection of the other ISFETs in the array, which is then coupled to analog ground VSSA as shown in FIG.

  In addition to the pixel design shown in FIGS. 9 and 24 (based on constant ISFET drain current and constant ISFET drain source voltage), p-channel ISFETs according to other inventive aspects of the present disclosure, as shown in FIGS. Alternative pixel designs for ISFET arrays based on both n-channel ISFETs are contemplated. As discussed below, some alternative pixel designs require additional and / or modified signals from the array controller 250 to facilitate data acquisition. In particular, a common feature of the pixel designs shown in FIGS. 25-27 is the inclusion of a sampling and holding capacitor within each pixel itself in addition to the sampling and holding capacitor for each column of the array. The alternative pixel design of FIGS. 25-27 generally includes more elements than the pixel design of FIGS. 9 and 24, but the pixel sampling and holding capacitor feature enables a “snapshot” type array. Where all pixels of the array are enabled simultaneously to sample a complete frame and obtain a signal representative of one or more analyte measurements proximate to each ISFET of the array. In some applications, this can provide higher data acquisition speed and / or improved signal sensitivity (eg, higher signal to noise ratio).

  FIG. 25 shows one such alternative design for one pixel 105C and associated column circuit 110j. Pixel 105C uses an n-channel ISFET and is generally based on the assumption that it provides a constant voltage across ISFET Q1 based on a feedback amplifier (Q4, Q5, and Q6). In particular, transistor Q4 constitutes the load of the feedback amplifier, and the amplifier current is set by a bias voltage VB1 (provided by the array controller). Transistor Q5 is a common gate amplifier, and transistor Q6 is a common source amplifier. Again, the purpose of the feedback amplifier is to keep the voltage across ISFET Q1 constant by adjusting the current supplied from transistor Q3. Transistor Q2 limits the maximum current that ISFET Q1 can draw (eg, this prevents damage caused by overheating of a very large array of pixels). This maximum current is set by the bias voltage VB2 (also provided by the array controller). In one aspect of the pixel design shown in FIG. 25, the power to pixel 105C can be turned off by setting bias voltage VB2 to 0V and bias voltage VB1 to 3.3V. In this way, the power delivered to the large array of such pixels can be adjusted (turned on for a short time and then turned off by the array controller), while simultaneously measuring the ion concentration while reducing the overall power consumption of the array. Can be obtained. Adjusting the power to the pixels can also reduce the heat dissipation of the array, reducing the possibility of heating the analyte solution, and thereby reducing the potentially harmful effects of sample overheating.

  In FIG. 25, the output of the feedback amplifier (gate of transistor Q3) is sampled by MOS switch Q7 and stored in pixel sampling and holding capacitor Csh in the pixel itself. Switch Q7 is controlled by a pixel sampling and holding signal pSH (supplied by the array controller to the array chip), which is applied simultaneously to all pixels of the array to read all pixels on their respective sampling and holding capacitors. Save at the same time. Thus, an array based on the pixel design of FIG. 25 is a “snapshot” array in that it samples all frames of data at any given time instead of sampling successive rows of the array. Conceivable. After each pixel value is stored in the corresponding pixel sampling and holding capacitor Csh, each pixel 105C (ISFET and feedback amplifier) may take another pH reading or be turned off to save power You can also

In FIG. 25, the pixel values stored in all the pixel sampling and holding capacitors Csh are applied to the column circuit 110j, one row at a time, through the source follower Q8, and the source follower receives the row selection signal (eg, Rowsel1). In response to the transistor Q9. In particular, after selecting and confirming a row, the value stored in the pixel sampling and holding capacitor is then stored in the column sampling and holding capacitor Csh2 enabled by the column sampling and holding signal COL SH, An output signal V _COLj is supplied.

FIG. 26 illustrates another alternative design for one pixel 105D and associated column circuit 110j according to one aspect of the present disclosure. In this embodiment, the ISFET is shown as a p-channel device. At the start of the data acquisition cycle, the CMOS switches controlled by signals pSH (pixel sampling / holding) and pRST (pixel reset) are closed (these signals are supplied by the array controller). This pulls the power supply of ISFET (Q1) to the voltage VRST. Subsequently, when the switch controlled by the signal pRST is opened, the power supply of ISFET Q1 pulls the pixel sampling and holding capacitor Csh to a threshold below the level set by pH. Next, the switch controlled by the signal pSH is opened and the pixel output value is coupled to the column circuit 110j via the operation of the switch in response to the row selection signal RowSel1 to provide the column output signal V _COLj . Similar to the pixel design in the embodiment shown in FIG. 25, the array based on pixel 105D is also a “snapshot” array in that it operates on all pixels of the array simultaneously. In one aspect, this design allows for a long and simultaneous integration time for all pixels, followed by high speed reading of all frames of data.

FIG. 27 illustrates another alternative design for one pixel 105E and associated column circuit 110j according to one aspect of the present disclosure. In this embodiment, the ISFET is again shown as a p-channel device. At the start of the data acquisition cycle, the switch operated by the control signals p1 and pRST is closed for a short time. Thereby, the value stored in the sampling capacitor Csh is cleared, and the charge can be stored in the ISFET (Q1). Next, the switch controlled by the signal pSH is closed to allow the charge stored in the ISFET Q1 to be stored in the pixel sampling and holding capacitor Csh. Next, the switch controlled by the signal pSH is opened, and the pixel output value is coupled to the column circuit 110j through the operation of the switch in response to the row selection signal _RowSel1 to provide the column output signal V _COLj . Pixel 105E is supplied with gain through the ratio of ISFET capacitance to Csh capacitance, ie, gain = C _Q1 / C _sh , or the pixel is enabled multiple times (ie, multiple samples of analyte measurement). And by accumulating the ISFET output to the pixel sampling and holding capacitor Csh without resetting the capacitor, gain is provided (ie, gain is a function of the number of accumulations). Similar to the embodiment of FIGS. 25 and 26, the array based on pixel 105E is also a “snapshot” array in that all pixels of the array can be manipulated simultaneously.

  Apart from sensor considerations, we now consider the combination of ISFET arrays with microwell arrays and associated fluids. Since most views of the microarray structure are only cross-sectional views or show the array only as a block in a simplified view, FIG. 28A and FIG. Help visualize in dimensional space. FIG. 28A shows a group of round cylindrical wells 2810 arranged in one array, and FIG. 28B shows a group of square columnar wells 2830 arranged in one array. It can be seen that the wells are separated (separated) from each other by the material 2840 forming the walls of the well. Obviously, it is possible to produce wells with other cross-sectional shapes, but this is not considered advantageous. Such an array of microwells is placed on top of the above-described ISFET array to accommodate one or more ISFETs per well. One can imagine one of these arrays when identifying the microarray in the following figure.

Fluid System: Apparatus and Method for Use with a High Density Electronic Sensor Array For many uses, an array element to complete a system for sensing chemical reactions or chemicals using the high density electronic array described above. Techniques and devices are needed to deliver fluids containing chemical or biochemical elements for sensing (referred to as “pixels”). In this section, exemplary techniques and methods are described, which have the desired characteristics and are useful for such purposes.

  Since high speed operation of the system will be desired, the fluid delivery system preferably does not limit the overall system operating speed as much as possible.

  Thus, the sample to be evaluated should be sensed as well as the need for a high speed and high density array of ISFETs or other elements that are sensitive to ion concentrations or other chemical attributes, or changes in chemical attributes. There is also a need for associated mechanisms and techniques for supplying the array elements in sufficiently small reaction doses to substantially improve the quality and speed of variable detection.

  There are two or even three elements or subsystems and related methods involved in delivering a chemical sample of interest to the array element: (1) a macrofluidic system of reagent and wash fluid supply and suitable Valve operation and auxiliary devices, (2) flow cells, and (3) microwell arrays in many applications. Consider each of these subsystems in reverse order.

As discussed elsewhere, for many uses such as DNA sequencing, it is desirable to provide a corresponding array of microwells on a semiconductor sensor array, where each microwell is Preferably it is small enough to receive only one DNA-loaded bead, in this regard, the pixels below the array provide a corresponding output signal.

  The use of such a microwell array involves three stages of manufacture and preparation, each of which is considered separately: (1) Create an array of microwells and have a coating that includes a microwell array layer Also create a chip, (2) attach the coated chip to the fluid interface, and for DNA sequencing, (3) load one or more DNA loading beads into the well. Of course, it will be appreciated that in other applications beads may be unnecessary or beads with different properties may be used.

Manufacture of microwell arrays Microwells may be manufactured by a number of methods. The actual details of manufacturing may require some experimentation and will vary depending on the processing capabilities available.

  In general, for the production of high-density arrays of microwells, the well array structure is photolithography using an etching method on one or more layers of materials such as photoresist (organic or inorganic), dielectrics, etc. Patterning. Patterning may be performed with the material on the sensor array, or may be performed separately and then transferred onto the sensor array chip, or some combination of the two. However, techniques other than photolithography should not be excluded if they give acceptable results.

  One example of forming a microwell array will now be considered, starting with reference to FIG. This figure shows a top view of one corner (ie, the lower left corner) of the chip layout showing one array 2910 of individual ISFET sensors 2912 on a CMOS die 2914. Signal lines 2916 and 2918 are used to address the array and read its output.

  Block 2920 represents several electronic devices for the array, as described above, and layer 2922 represents a portion of the wall, which is a portion of a microfluidic structure or flow cell, as described in detail below. It becomes. The flow cell is a structure that provides fluid flow either above the microwell array or directly on the sensor surface if there is no microwell structure. On the surface of the die, a pattern such as the pattern 2922 in the lower left of FIG. 29 can be formed during semiconductor processing to form the ISFET and related circuitry when the dielectric covers the die surface. , Can be used as alignment marks to place wells on sensor pixels.

  After forming the semiconductor structure as shown, the microwell structure is applied to the die. That is, the microwell structure can be formed directly on the die, or can be formed separately and then mounted on the die, either approach being acceptable. Various methods may be used to form the microwell structure on the die.

  For example, the entire die may be spin coated to the desired height of the microwell with a negative photoresist such as Microchem's SU-8 2015 or a positive resist / polyimide such as HD Microsystems HD8820. The desired height of the well in one or more photoresist layers (e.g., but not limited to about 4-12 [mu] m in the example of one pixel per well) is one or more layers In, a suitable resist can be realized by spinning at a predetermined speed (this can be found by reference to the literature and according to the manufacturer's specifications or experimentally).

(Well height may typically be selected in relation to the lateral dimensions of the sensor pixel, preferably by a nominal 1: 1 to 1.5: 1 aspect ratio, height: width or diameter. Based on signal-to-noise considerations, there is a relationship between dimensions and the required data sampling rate to achieve the desired level of performance, so select the optimal parameters for a given application. There are a number of factors to consider.) Alternatively, multiple layers of different photoresists can be applied, or other forms of dielectric material may be deposited. Various types of chemical vapor deposition methods can be used to construct material layers suitable for microwell formation.

  After a photoresist layer (including the singular “layer” term, including multiple layers in the assembly) in place, individual wells (typically either one or four per well) Are mapped onto having a mask (eg, a chromium mask) on a resist-coated die and the resist is exposed to cross-linking (typically UV) radiation. The entire resist exposed to irradiation (ie, the portion of the mask that does not block irradiation) is cross-linked, resulting in a permanent plastic layer bonded to the surface of the chip (die). Non-reactive resist (ie, areas of the resist that were not exposed because the mask blocked the light from reaching the resist and prevented crosslinking), for example, propylene glycol methyl ethyl acetate (PGMEA) or other suitable It is removed by washing the chip in a suitable solvent (ie developer) such as a solvent. The resulting structure defines the walls of the microwell array.

  FIG. 30 shows an example of part of the layout of the chrome mask 3010 for one sensor / well aspect corresponding to part of the die shown in FIG. Gray regions 3012 and 3014 are regions that block UV irradiation. The alignment mark 3016 in the white portion of the lower left quadrant of FIG. 30 within the gray region 3012 is used to align the well layout with the ISFET sensor on the chip surface. A circular array 3014 in the upper right quadrant of the mask prevents irradiation from reaching the well region, leaving unreacted resist that can be dissolved in forming the well.

  FIG. 31 shows the corresponding layout of the mask 3020 for the 4 sensor / 1 well embodiment. The alignment pattern 3016 is still in use, and the individual well mask circle 3014A in the array 2910 now has twice the diameter of the well 3014 of FIG. 30 and instead of 1 sensor / 1 well. 4 sensors / well.

  After exposing the die / resist to UV radiation, a second layer of resist may be coated on the chip surface. This resist layer may be relatively thick and is typically about 400-450 μm thick. The second mask 3210 (FIG. 32), which may also be made of chrome, is used to mask the region 3220 surrounding the array, resist collar or wall 3310 (or basin, this The term is used in a geological sense), which surrounds an active array of sensors on a substrate 3312, as shown in FIG. In the particular example described, the collar is 150 μm wide in the x direction on each side of the array and 9 μm wide in the y direction on each side of the sensor array than the sensor array. The alignment marks (mostly not shown) of the mask 3210 match up with the alignment marks on the first layer and the CMOS chip itself.

  Other photolithographic approaches may be used to form the microwell array and, of course, the above is only an example.

  For example, contact lithography of different etchants and developers with different resolutions may be used. Both organic and inorganic materials may be used for one or more layers forming the microwell. One or more layers may be etched on a chip having a dielectric layer over the pixel structure of the sensor array, for example a passivation layer, or one or more layers may be formed separately, Next, it may be applied to a sensor array. The specific choice or method will depend on factors such as the size of the array, the size of the wells, available manufacturing facilities, acceptable cost, and the like.

  Among the various organic materials that may be used in some embodiments for forming one or more microwell layers are the SU-8 negative working photoresists described above, conventional positive working photoresists. And positive acting photosensitive polyimide. Each of these has advantages and disadvantages and is well known to those skilled in the photolithography art.

Of course, modifications are reasonable in the manufacturing environment.
Contact lithography is limited and may not be the manufacturing method selected to produce the highest density wells, i.e., in the lateral direction may result in values greater than the desired minimum pitch limit.

  Other techniques such as deep ultraviolet step-and-repeat can provide higher resolution lithography and can be used to produce smaller pitches and potentially smaller well diameters. Of course, different techniques will prove optimal for different desired specifications (eg number of sensors and wells per pitch). Practical factors such as manufacturing methods available to manufacturers also motivate the use of specific manufacturing methods. Although new methods are being considered, various aspects of the invention are limited to the use of these new methods.

  Preferably, the CMOS wafer with ISFET array is planarized after the final metallization process. Chemical mechanical dielectric planarization prior to silicon nitride passivation is preferred. This allows subsequent lithography steps to be performed on a very flat surface without back-end CMOS topography.

  A deep UV step-and-repeat lithography system can be utilized to resolve small features with excellent resolution, registration, and repeatability. However, the high resolution and large numerical aperture (NA) of these systems prevent them from having a large depth of focus. Therefore, when using such a manufacturing system, pattern transfer is performed using a thinner photosensitive spin-on layer (ie, a resist on the order of 1 to 2 μm rather than the thick layer used in contact lithography), and then the underlying 1 or It may be necessary to etch microwell features in more than one layer. For example, four 1 μm plasma chemical vapor deposition (standard manufacturing method) can be performed in sequence to provide the desired 4 μm microwell thickness. Next, pattern the features of the microwell using high resolution lithography and use a conventional SiO2 chemical etch with selective etch stop-once for the bond pad area and then once for the microwell area. Can do. The etch stop can be performed on an aluminum bond pad and silicon nitride passivation (or similar), respectively. Alternatively, other suitable surrogate pattern transfer and etching methods can be used to provide microwells of inorganic material.

  Another approach is to form a microwell structure in the organic material. For example, using a double resist “soft mask” method, a thin high-resolution deep UV resist can be used on top of a thick organic material (eg, cured polyimide or an opposite acting resist). The top resist layer is patterned. The pattern can be transferred using an oxygen plasma reactive ion etching method. This process sequence is sometimes referred to as “Portable Conformable Mask (PCM)” technology. BJ Lin et al., “Practicing the Novolac deep-UV portable conformable masking technique”, Journal of Vacuum Science and Technology 19, No. 4, 1313-1319 (1981) and A. Cooper et al, “Optimization of a reactive spin -on dielectric process for copper inductor coil and interconnect protection in RF SoC devices ”.

  Alternatively, a “drill-focusing” technique may be used, which limits high resolution steppers when patterning thick resist layers with several step-and-repeat exposures at different depths of focus. Compensated for depth of focus (DOF). This technique also depends on the NA and DOF of the stepper and the contrast properties of the resist material.

  Other PCM techniques can be applied for this purpose, such as those shown in US Patent Publication No. 2006/0073422 by Edwards et al. This is a three-layer PCM method and is shown in FIG. 33A. As shown here, a microwell array is basically produced by six major steps, and the results are very similar to those brought about by contact lithography.

  In a first step 3320, a layer of high contrast negative working photoresist of the type, such as Shipley InterVia Photodielectric Material 8021 (IV8021) 3322, is spun onto the surface of the wafer and this is the substrate 3312 of FIG. Assuming the wafer providing the sensor array here), a soft printing operation is performed. Next, in step 3324, a blocking antireflective coating (BARC) layer 3326 is applied and soft baked. On top of this structure, a thin resist layer 3328 is spun and soft baked, and in step 3330 a thin layer of resist is suitable for defining fine features. Resist layer 3328 is then patterned, exposed and developed, and the BARC of exposed areas 3329 not protected by resist 3328 is removed in step 3332. This opens region 3329 up to the uncured IV8021 layer. The BARC layer can now act like a conformal contact mask. Blanket exposure with the flood exposure tool in step 3334 cross-links the exposed IV8021, which is shown here as distinguished from uncured IV8021 at 3332. One or more development steps 3338 are then performed to remove all but the bridge IV8021 in region 3336. Region 3336 now constitutes the wall of the microwell.

  As shown above, the well is at the lowest position above the ISFET passivation layer (i.e. ends here), but the improvement in ISFET sensor performance (i.e. signal to noise ratio, etc.) is It is believed to be obtained when the active bead (s) is maintained slightly above the ISFET passivation layer. One way to achieve this is to place spacer “bumps” within the boundaries of the microwell pixels. One example of how to do this is to not etch away a portion of one or more layers used to form the microwell structure (ie, forming the microwell in two lithography steps). -1 time etching halfway, and the other time patterning the bumps and etching to the end) by depositing separate layers, defining by lithography and etching to "bump" Or by using a permanent photosensitive material for the bumps once the microwell has been completed. An alternative (or additional) non-integrated approach is to load the wells with one or more layers of very small packing beads before loading the DNA-carrying beads.

Mounting coated chip on flow cell (fluid interface)
The process of sequencing the DNA in a sample using an array of sensor arrays on a chip in combination with an array of microwells is called an “experiment”. Performing the experiment requires loading the wells with DNA binding beads and flowing several different solutions (ie, reagents and wash solutions) throughout the well. There is a need for a liquid delivery system coupled to a fluid interface, which is a dead volume that is small enough to allow various solutions throughout the well, with controlled laminar flow, and low secondary contamination between successive solutions. Have a shed. The fluid interface is sometimes referred to as a “flow cell”.

Many configurations of flow cell designs are possible. The systems and methods presented herein do not depend on the use of a specific flow cell configuration. However, a suitable flow cell substantially fits the following set of objectives:
-A suitable interconnection with the fluid delivery system-eg via appropriately sized tubing.
• Appropriate headspace above the well (about 300 μm for dna sequencing).
Minimize dead volume (ie, the space contained above the microwell array) that fluid encounters before entering the flow chamber.
-Removal of small spaces in contact with liquid, but not through the flow cell (to minimize cross-contamination).

• Uniform expansion of the flow from the inlet piping towards the wide / flat front at the entrance to the flow chamber.
• Laminar flow characteristics are such that they maintain a wide / flat front profile while traversing the entire chip from the inlet side to the outlet side.
• be compatible with the placement of the removable reference electrode inside or as close as possible to the flow chamber.
-Easy loading of beads.
• It can be manufactured at an acceptable cost.
-Easy assembly of the flow cell and attachment to the chip package.

  Each of several example designs will be considered to meet these criteria. In each example, one would generally choose to implement the design using one of two methods. The flow cell is attached to the frame and the frame is adhered (or otherwise attached) to the chip, or the frame is integrated into the flow cell structure and the integrated assembly is attached to the chip. In addition, designs can be categorized by how the reference electrode is incorporated into the arrangement. Depending on the design, the reference electrode can be integrated into the flow cell (eg, forming part of the ceiling of the flow chamber) or can be placed in the flow path (generally the outlet of the flow path). Or downstream, after the sensor array).

  A first example of a suitable experimental apparatus 3410 incorporating such a fluid interface is shown in FIGS. The manufacture and structure of this will be discussed in more detail below.

  The apparatus includes a semiconductor chip 3412 (generally indicated but hidden), on or in which an array of wells and sensors is formed, and further includes a fluid assembly 3414 on the chip; This delivers the sample to the chip for reading. The fluid assembly includes a portion 3416 for introducing a fluid containing a sample, a portion 3418 for draining the fluid through a tube, and a flow chamber that allows fluid to flow from the inlet to the outlet and interact with the material in the wells along the way. A portion 3420 is included. These three parts include a glass slide 3422 (eg, Erie Microarray Cat # C22-5128-M20 from Erie Scientific Company, Portsmouth, NH cut to one third of a 25 mm × 25 mm size). Integrated by interface.

  There are two mounting parts 3424 and 3426 on the top surface of the glass slide, for example the nanoport mounting part Part # N-333 from Upchurch Scientific of Oak Harbor, WA. One port (eg, 3424) serves as an inlet to deliver liquid from the pump / valve operating system described below (but not shown). The second port (eg 3426) is the outlet and flows through the tube to discard the liquid. Each port is connected to a conduit 3428, 3432 such as a flexible tubing of appropriate inner diameter. The nanoport is mounted so that the pipe can penetrate the corresponding hole in the glass slide. The opening of the tube is flush with the bottom surface of the slide.

  At the bottom of the glass slide, the flow chamber 3420 may include various structures to facilitate substantial laminar flow throughout the microwell array. For example, a series of microfluidic channels extending from the inlet tube to the end of the flow chamber can be patterned by contact lithography with a positive photoresist such as SU-8 photoresist from MicroChem Corp. of Newton, MA. Good. Other structures are considered below.

Chip 3412 is then mounted on carrier 3430 for packaging and connection with connector pins 3432.
For ease of description, the glass slide 3422 will now be considered upside down compared to the arrangement in FIGS.

  A layer of photoresist 3810 is applied to the “top” of the slide (this is the “bottom” when the slide and the added layer are inverted and attached to the ISFET array sensor assembly with the microarray mounted). Layer 3810 may be about 150 μm thick in this example, and forms a first fluid transport layer from the end of the nanoport tubing to the end of the sensor array chip. Layer 3810 is patterned using a mask, such as mask 3910 of FIG. 39 ("patterning" refers to exposing the resist through the mask using a radiation source and then removing the unplasticized resist. means). The mask 3910 has an irradiation transmission area indicated by white and an irradiation block area 3920 indicated by shading.

  The irradiation block region is 3922 to 3928. Region 3926 forms a channel around the sensor assembly. This is formed about 0.5 mm inward from the outer boundary of the mask 3920, thus avoiding typical edge beads. Regions 3922 and 3924 block the irradiation, thereby removing corresponding portions of the resist to form voids in the shape shown. Each of the regions 3922 and 3924 has a rounded end dimensioned to receive one corresponding end of a tube 3428, 3432 that passes through a corresponding nanoport 3424, 3426. From the rounded end, the regions 3922, 3924 extend in the direction of the sensor array, allowing the liquid to spread and allowing the flow through the array to be substantially laminar. Region 3928 is a simple alignment pattern, may be any suitable alignment pattern, or may be replaced by a suitable alternative alignment mechanism. The dashed lines in FIG. 38 are provided to illustrate the formation of voids 3822 and 3824 under mask regions 3922 and 3924.

  The second layer of photoresist is formed quite separately rather than on resist 3810 or slide 3422. Preferably, this is formed on a flat flexible surface (not shown) to produce a peelable patterned plastic layer. This second layer of photoresist can be formed using a mask, such as mask 4010, which leaves the following on the flexible substrate after patterning: boundary, region under region 4012, region Alignment marks made by two slits below 4014, 4016 (where use is discussed below), and patterning regions 4018 and 4022. A second layer of photoresist is then applied to the first layer of photoresist, but this time for alignment of these layers, for example if created with pattern 4018, one alignment mark or This is done using a set of alignment marks. The second layer is then peeled from the flexible substrate and the latter is removed.

Using another alignment mark or set of alignment marks produced by pattern 4022, alignment with the next layer, discussed below, is performed.
The second layer is preferably about 150 μm deep, which covers the fluid transport channel, but is about 150 μm at each end of the sensor array chip below the slit-forming regions 4014 and 4016. Excluding length slits.

  After the second layer of photoresist is disposed on the first layer, a third photoresist patterning layer is formed on the second layer using a mask such as mask 4110 shown in FIG. . The third layer provides a baffle element under the region 4112, which is the same width as the collar 3310 of the sensor chip array (see FIG. 33), but overlaps the fluid transport channel of the first layer. To make it possible, it is narrowed by about 300 μm.

  The third layer is about 150 μm thick and penetrates the tip collar 3310 by about 150 μm towards the basin floor formed thereby. This structure leaves a headspace of about 300 μm above the wells of the sensor array chip. The liquid is flowed through the 150 μm slit under 4014, 4016 along the entire width of the sensor array across the well.

  FIG. 36 is a partial cross-sectional perspective view of the above example embodiment of a microfluidic and sensor assembly, which is also shown in FIGS. 34 and 35, but in FIG. It has been enlarged for visual recognition. (An enlarged schematic view of half of the flow path is shown in FIG. 37.) It can now be seen that the fluid enters through the inlet tube 3428 of the inlet port 3424. At the bottom of tube 3428, fluid flows through fluid expansion chamber 3610 formed by mask region 3922, and fluid passes over collar 3310 and then down to basin bottom 3320, across die 3412 and its microarray. Flowing. After passing over the array, the fluid then bends at a right angle at the far wall of the collar 3310 and over the collar across the flow concentration chamber 3612 formed by the mask region 3924 and through the outlet tube 3432 of the outlet port 3426. It is discharged outside through. A portion of this flow from the middle of the array to the outlet can also be seen in the enlarged view of FIG. 37, where the arrows indicate the fluid flow.

The fluid assembly can be secured to the sensor array chip assembly by applying an adhesive to the mating surface portions of the two assemblies, aligning and pressing them together.
Although not shown in FIGS. 34-36, it is understood that the reference electrode is a metallization 3710 in the ceiling of the flow chamber, as shown in FIG.

  Another method of introducing a reference electrode is shown in FIG. Here, a hole 4210 is provided in the ceiling of the flow chamber and a grommet 4212 (eg, made of silicone) is fitted into the hole to provide a central passage or bore into which the reference electrode 4220 can be inserted. Baffles or other micro features (not shown) may also be patterned into the flow channel to facilitate laminar flow over the microwell array.

  FIGS. 43-44 illustrate another alternative flow cell design 4310. FIG. This design relies on a single plastic piece or element 4320 molding method to attach to the chip and complete the flow cell. The connection to the fluid system is made via a threaded connection, which is placed in appropriate holes in the plastic piece at 4330 and 4340. Alternatively, if the element 4320 is made from a material such as polydimethylsiloxane (PDMS), the connection may be made by simply inserting the tubing into an appropriately sized hole in the element 4320. Vertical cross-sectional views of this design are shown in FIGS.

  This design uses an overhanging plastic collar 4350 (which may be a series of spaced apart legs forming a solid wall as shown or a fence-like wall extending downwards). Thus, the chip package may be included and the plastic piece aligned with the chip package, or other suitable structures may be used, thereby aligning the chip frame with the flow cell forming element 4320. Liquid is directed into the flow cell through one of the openings 4330, 4340 and from there downwards into the flow chamber.

  In the illustrated embodiment, the reference electrode is introduced through the bore 4325 of the element 4320 at the top of the flow chamber. Placement of the removable reference electrode is facilitated by a silicone sleeve 4360 and an epoxy stop ring 4370 (see enlarged view in FIG. 44). The silicone sleeve provides a seal and the epoxy stop ring prevents the electrode from being inserted too far into the flow cell. Of course, other mechanisms may be used for the same purpose, and a structure for stopping the electrode does not have to be used. Also, when a material such as PDMS is used for the element 4320, the material itself forms a waterproof seal when the electrode is inserted, eliminating the need for a silicone sleeve.

  45 and 46 are similar arrangements, except that element 4510 does not have a bore for receiving a reference electrode. Instead, the reference electrode 4515 is formed at or attached to the bottom of the central portion 4520 and forms at least a portion of the ceiling of the flow chamber. For example, a metallization layer may be applied to the bottom of the central portion 4520 and before the element 4510 is attached to the chip package.

  FIGS. 47-48 show another example, which is a variation of the embodiment shown in FIGS. 43-44, where the frame is manufactured as part of a flow cell and before the frame is attached to the chip surface. The flow port structure is not attached to the frame. In this type of design, the assembly is somewhat more precise because the wire bond to the chip is not protected by the epoxy encapsulating the chip. The success of this design depends on the precise placement and secure adhesion of the integrated “frame” to the chip surface. FIGS. 49-50 show an embodiment corresponding to FIGS. 45-46 in which the reference electrode 4910 is on the ceiling of the flow chamber and the frame is manufactured as part of the flow cell.

  Yet another alternative to the fluid assembly has a fluid element 5110 on the standoff 5120 raised about 5.5 mm above the top surface of the chip package 5130, as shown in FIGS. This allows the operator to visually inspect the quality of the bond between the plastic piece 5140 and the chip surface and externally strengthen the bond if necessary.

  Some of the aforementioned alternative aspects may also be implemented in a hybrid plastic / PDMS configuration. For example, as shown in FIGS. 53-54, plastic portion 5310 forms a frame and flow chamber and rests on PDMS “base” portion 5320. Plastic portion 5310 also provides region 5330 for the array for expansion of fluid flow from the inlet port. The PDMS section may also include communication slits 5410, 5412 through which liquid passes from or to the lower flow chamber.

  The fluid structure may also be made of glass as described above, eg, photosensitive (PD) glass. Once such glass has been selectively exposed to UV light, it has an enhanced etch rate in hydrofluoric acid, which allows it to be micromachined to the top and back surfaces when it is bonded together. A three-dimensional low aspect ratio fluid cell can be formed.

  An example is shown in FIG. The first glass layer or sheet 5510 is patterned and etched to create nanoport fluid holes 5522 and 5524 on the top surface and fluid expansion channels 5526 and 5528 on the back surface. The second glass layer or sheet 5530 is patterned and etched to provide downward fluid input / output channels 5532 and 5534 at a height (layer thickness) of about 300 μm. The bottom surface of layer 5530 is thinner toward the outside of channels 5532 and 5534, allowing layer 5530 to rest on the chip frame and allow protrusion 5542 to form the top surface of the flow channel at the appropriate height. . Two glass layers, or wafers, and four lithography steps are required. Both wafers are aligned and bonded (eg, with a suitable adhesive not shown) so that the downward fluid input / output ports are properly aligned with the fluid expansion channels. The alignment target can be etched into the glass to facilitate the alignment process.

The nanoport can be secured over the nanoport fluid hole to facilitate connection of input and output piping.
The central bore 5550 can be etched through the glass layer to receive the reference electrode 5560. The electrode can be fixed and sealed in place by a silicone collar 5570 or similar structure. Alternatively, the electrode can be constructed integrally with a suitable washer to enable the same purpose.

  By using a glass material in a two-layer fluid cell, the reference electrode may be a conductive layer or pattern deposited on the bottom surface of a second glass layer (not shown). Alternatively, as shown in FIG. 56, the projecting region is etched to form a transmissive glass film 5610, and a silver (or other material) thin film 5620 is coated thereon to form an integrated reference electrode. Hole 5630 is etched into the upper layer for access to the electrode, and if the hole is large enough, it can also function as a reservoir for the silver chloride solution. The electrical connection with the thin film silver electrode may be made in any suitable manner, such as by wire bonding to a clip-on pushpin connector, or alternatively to a ceramic ISFET package.

  Yet another exemplary embodiment of a fluid assembly is shown in FIGS. This design is used to connect the plastic piece 5710 incorporating the frame and directly attached to the chip surface, and the piping from the fluid system, and similar to the PDMS piece described above, the liquid is placed in a small bore tube. To the second piece 5720, which is dispersed into a wide flat slit.

  The two pieces are glued together and by using multiple (eg, three) alignment markers (not shown), the two pieces can be accurately aligned during the gluing process. Holes can be provided in the bottom plate, which holes are used to fill the void with epoxy (example), protect the wire bond to the chip, and fill any potential void in the frame / chip contact. In the example shown, the reference electrode is external to the flow cell (through the outlet port, downstream of the exhaust stream, see below), although other structures of the reference electrode may of course be used.

  Still another example of the flow cell structure is shown in FIGS. 59A includes eight views (AH) of an injection molded bottom layer or plate 5910 for a flow cell fluid interface, and FIG. 59B shows seven types of mating injection molded top plate or layer 5950. Includes figures (A-G). The bottom of the plate 5910 has a downward dependent rim 5912 positioned to surround the sensor chip and an upwardly extending rim 5914 for mating with the top plate 5610 along its outer edge. Thus, two liquid chambers (an inlet chamber and an outlet chamber) are formed between them. A stepped downward dependent portion 5960 of the upper plate 5950 separates the inlet chamber from the outlet chamber. The inlet tube 5970 and the outlet tube 5980 are formed integrally with the rest of the upper plate 5950. From the inlet tube 5970 that pours at the small end of the inlet chamber formed by the depression 5920 at the top of the plate 5910, it spreads out to the outlet end of the inlet chamber and fluid is directed across the entire array.

  Even if the flow cell is formed using glass or plastic or other materials, especially for large arrays, there is simply a gradually expanding (spreading) space in the flow cell inlet chamber, between the inlet line and the front end of the array. Rather, it may be desirable to include a structure that facilitates the flow through the array to be a suitable laminar flow. An example of a structure for this purpose, shown in FIG. 59C, using the bottom layer 5990 of an injection molded flow cell as an example is the tree structure 5992 of the channel from the flow cell inlet position to the front end of the microwell array or sensor array. This is understood at 5994 below the outlet side of the structure.

  There are various other ways to provide a fluid assembly for delivering the appropriate fluid flow through the microwell and sensor array assembly, and the foregoing examples are therefore not intended to be exhaustive.

Reference electrodes Commercially available flow-type fluid electrodes, such as silver chloride proton permeable electrodes, can be inserted into the fluid system in series, and these electrodes are generally aligned with the fluid system for various electrochemical purposes. Designed to provide a stable potential. However, in the above system, such a potential must be maintained in the fluid volume in contact with the microwell ISFET chip. With conventional silver chloride electrodes, it has been found difficult to achieve a stable potential because of the electrically long fluid path (through a small channel in the flow cell) between the chip surface and the electrode. This results in the reception of noise in the chip electronics. Furthermore, the large volume within the flow cavity of the electrode tends to trap and accumulate bubbles that degrade the electrical connection with the fluid.

  Referring to FIG. 60, a solution to this problem is that of a stainless steel capillary electrode 6010 connected directly to the outlet port 6020 of the chip flow cell and connected to a power source (not shown) via a shielded cable 6030. Found in use. The metal capillary 6010 has a small inner diameter (eg, about 0.01 ") and effectively transports fluids and gases, as well as other microfluidic piping, without confining the gas to an appreciable extent. Can be inserted directly into the flow cell port 6020, so it is close to the chip surface and reduces potential electrical loss through the fluid. The large internal surface area of the capillary (typically about 2 "long) It can contribute to high performance. The construction with stainless steel is highly chemically resistant and does not produce electrochemical effects due to its use at very low currents (<1 μA) in the system. A fluid attachment 6040 is attached to the end of the capillary that is not in the flow cell port for connection to tubing to the fluid delivery and removal subsystem.

A complete system for using a fluid system sensor array will include a controller for valve operation for low reagent and wash solutions on the microarray or sensor array, depending on the preferred fluid source, valve operation and application. Let's go. These elements can be easily assembled from off-the-shelf elements, and the controller can be easily programmed to perform the desired experiment.

  As already discussed, the devices and systems of the present invention can be used to detect and / or monitor interactions between various entities. These interactions include chemical or enzymatic reactions in which substrates and / or reagents are consumed and / or reaction byproducts are generated. An example of an enzymatic reaction that can be monitored according to the present invention is nucleic acid sequencing, which is discussed in detail herein. The devices and systems provided herein in the context of sequencing reactions can detect nucleotide incorporation based on changes in chemFET current.

  The current change may be the result of a single or several combinations of one or more of the following events: production of PPi, production of Pi (eg in the presence of pyrophosphatase), production of hydrogen ( Simultaneous changes in pH and pH (for example, in the presence of low-strength buffers), reduced concentration of unincorporated dNTPs on the chemFET surface, delayed arrival of dNTPs on the chemFET surface, etc. It should be understood that the methods provided herein are independent of the mechanism that causes the current change. Thus, the present invention contemplates performing nucleic acid sequencing based on changes in chemFET current. The methods provided herein in connection with sequencing can be contrasted with methods in the literature including Pourmand et al. PNAS 2006 103 (17): 6466-6470.

  FIG. 61 illustrates the production of PPi resulting from incorporation of nucleotides into a newly synthesized nucleic acid strand. PPi produced as a reaction by-product of nucleotide incorporation into the nucleic acid strand can be obtained in the absence of PPi receptor (eg, as shown in FIG. 11B) and in the absence of detectable pH change (eg, this Even those defined in the specification, such as those occurring in the presence of strong buffers, can be detected directly. In some cases, simply the presence of PPi causes an electrical change on the chemFET surface, which causes a current change. The current change may occur solely due to PPi generation, or may occur in combination with other events, such as those described above.

  Accordingly, in one aspect, the present invention contemplates nucleic acid sequencing using a chemFET array, such as an ISFET array. The method of the invention is a “sequencing by synthesis” method because it requires the synthesis of a new nucleic acid strand that is complementary to the strand to be sequenced.

  The release of PPi after incorporation of nucleotides into the newly synthesized nucleic acid strand is shown in FIG. Incorporation of dNTP into the nucleic acid strand releases PPi, which can then be hydrolyzed into two orthophosphates (Pi) and one hydrogen ion. Thus, the generation of hydrogen ions can facilitate the detection of nucleotide uptake based on pH changes. Alternatively, as described herein, PPi production (detected in the presence or absence of PPi receptor) can facilitate the detection of nucleotide incorporation based on pH changes. In still other embodiments, PPi can be converted to Pi using pyrophosphatase, and Pi can be detected directly or indirectly. Any or all of these events (and also those described herein) can be involved in causing a current change in chemFET that correlates with nucleotide incorporation.

  The sequencing reaction maximizes complete uptake across all wells for any given dNTP, reduces or reduces the number of unincorporated dNTP remaining in the well, and produces the highest possible signal-to-noise ratio. The goal is to achieve.

  The nucleic acid to be sequenced is referred to herein as the target nucleic acid. Target nucleic acids include, but are not limited to: DNA, such as, but not limited to, genomic DNA, mitochondrial DNA, cDNA, and RNA, and RNA, such as, but not limited to, mRNA, miRNA, and the like. The nucleic acid may be from any source, including natural or synthetic sources. Nucleic acids may be PCR products, cosmids, plasmids, natural or synthetic libraries, and the like. The present invention is not intended to be limited in this respect. The methods provided herein can be used for sequencing nucleic acids of any length. Specifically, a proof of principle demonstration of the sequencing of four templates of known sequences in the examples is provided. This artificial model is intended to show that the device and system can read nucleic acid uptake correlated with a known sequence of the template. This is not intended to indicate general usage in the art of the method or system. The following is a brief description of these methods.

  The target nucleic acid is prepared using any method well known in the art. As an example, genomic DNA is collected from a sample according to techniques well known in the art (see, eg, Sambrook et al. “Maniatis”). After collection, the DNA is fragmented to produce short length nucleic acids. The resulting fragments may be on the order of hundreds, thousands, or tens of thousands of nucleotides. In some embodiments, the fragment is 200-1000 base pairs in size, or 300-800 base pairs in size, but is not limited thereto. Nucleic acids can be fragmented by any method including, but not limited to, mechanical, enzymatic or chemical methods. Examples include shearing, sonication, nebulization, and endonuclease (eg, Dnase I) digestion or any other technique known to produce nucleic acid fragments, preferably of the desired length. including. After fragmentation, size selection methods can be used to enrich or isolate fragments of a particular length or size. Such techniques are also known in the art and include, but are not limited to, gel electrophoresis or SPRI.

  In some embodiments, a size-selected target nucleic acid is linked to both 5 'and 3' end adapter sequences. These adapter sequences include amplification primer sequences for use in amplification of the target nucleic acid. One adapter sequence also includes a sequence complementary to the sequencing primer. The opposite adapter sequence may include a moiety that facilitates binding of the nucleic acid to a solid support, such as but not limited to a bead. An example of such a moiety is a biotin molecule (or a double biotin moiety as described in Diehl et al. Nature Methods, 2006, 3 (7): 551-559), and such labeled nucleic acids are therefore avidin or It can be bound to a solid support having streptavidin groups. The resulting nucleic acid is referred to herein as a template nucleic acid. The template nucleic acid includes at least a target nucleic acid and usually includes a nucleotide sequence in addition to the target.

  In some cases, a spacer is used to provide a distance between the template nucleic acid (and particularly the target nucleic acid sequence contained therein) and the bead. This facilitates sequencing of the end of the target closest to the bead. Examples of suitable linkers are known in the art (see Diehl et al. Nature Methods, 2006, 3 (7): 551-559) and include, but are not limited to, carbon-carbon linkers such as iSp18.

The solid support to which the template nucleic acid binds is referred to herein as a “capture solid support”. Where the solid support is a bead, such bead is referred to herein as a “capture bead”. The beads can be made from any material, including but not limited to: cellulose, cellulose derivatives, gelatin, acrylic resins, glass, silica gel, polyvinylpyrrolidine (PVP), vinyl and acrylamide copolymers, polystyrene , Polystyrene cross-linked with divinylbenzene, etc. (see Merrifield Biochemistry 1964, 3, 1385-1390), polyacrylamide, latex gel, dextran, cross-linked dextran (eg Sephadex ^™ ), rubber, silicon, plastic, nitrocellulose, natural sponge , Metal, and agarose gels (Sepharose ^™ ). In one embodiment, the beads are streptavidin coated beads. The bead diameter depends on the density of the ISFET, and well arrays used with larger arrays (and thus smaller wells) require smaller beads. Generally, the bead size is about 1-10 μM, more preferably 2-6 μM. In some embodiments, the beads are about 5.91 μM, while in other embodiments, the beads are about 2.8 μM. It should be understood that the beads may or may not be completely spherical. It should be understood that other beads may be used and other mechanisms for attaching nucleic acids to the beads may be used.

  As discussed herein, the sequencing reaction is performed in a well located over the chemFET. Wells (here used interchangeably with reaction chambers or microwells) may vary in size between arrays. Preferably, the well width to height ratio is from 1: 1 to 1: 5. The bead-to-well dimensions are preferably in the range of 0.6 to 0.8.

  A homogeneous population of amplified nucleic acids is joined to one or more beads, where each bead ultimately binds to a plurality of identical nucleic acid sequences. The loading of the nucleic acid template onto the beads depends on a number of factors including the bead size and the length of the nucleic acid. In most aspects, maximum loading on the beads is desirable. Nucleic acid amplification and binding to solid supports such as beads is accomplished in a number of ways, such as described by Margulies et al. Nature 2005 437 (15): 376-380 and accompanying supplemental material. Including, but not limited to, emulsion PCR. In some embodiments, the amplification is representative amplification. A representative amplification is an amplification that does not change the relative presentation of any nucleic acid species.

  Before and / or during placement in the well of the flow chamber, the beads are complementary sequences located at the 3 ′ end of the template nucleic acid (ie, in the amplification primer sequence or other linked to the 3 ′ end of the target nucleic acid). Sequencing primer and polymerase that bind to the adapter sequence) and incubate for a time and under conditions that promote hybridization of the primer with its complementary sequence and binding of the polymerase to the template nucleic acid. A primer can be virtually any sequence as long as it is long enough to be unique. Hybridization conditions are such that the primer hybridizes only with its true complement at the 3 'end of the template. Suitable conditions are disclosed in Margulies et al. Nature 2005 437 (15): 376-380 and accompanying supplemental material.

  Suitable polymerases include, but are not limited to, DNA polymerases, RNA polymerases, or subunits thereof, as long as they are capable of synthesizing new nucleic acid strands based on templates starting from hybridized primers. An example of a suitable polymerase subunit is the exo form of the Klenow fragment of E. coli DNA polymerase I lacking 3 'to 5' exonuclease activity. Thus, the enzyme binds to the bead (or corresponding solid support) but not to the ISFET surface itself. Template nucleic acids can also be added to other reagents and / or cofactors including but not limited to buffers, detergents, reducing agents such as dithiothreitol (DTT, Cleland reagent), single-stranded binding proteins, etc. Contact before entering and / or while in the well. In one embodiment, the template nucleic acid is contacted with a primer and a polymerase prior to introducing it into the flow chamber and its wells.

  Nucleic acid loaded beads are introduced into the flow chamber and ultimately into wells located on the ISFET array. This method requires that each well in the flow chamber contains only one nucleic acid-loaded bead because, when there are two beads in one well, it comes from two different nucleic acids. This results in two unusable sequencing information. The example provides a brief description of one exemplary bead loading protocol in the context of magnetic beads. It should be understood that other bead types can be loaded using a similar approach. The protocol reduces the likelihood and incidence of air in the flow chamber wells, distributes nucleic acid loaded beads evenly throughout the flow chamber wells, and eliminates the presence and / or accumulation of excess beads in the flow chamber. It has been demonstrated to avoid.

  The percent occupancy of the wells on the chip can vary depending on the method being performed. A higher occupancy is desirable when the method aims to extract the maximum sequence data in the shortest possible time. Low occupancy is acceptable if speed and throughput are not as important. Thus, depending on the embodiment, a suitable occupancy percentage is at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90% of the wells. %, Or at least 100%. As used herein, occupancy refers to the presence of one nucleic acid loaded bead in a well, and occupancy percentage refers to the percentage of the well on the chip that is occupied by a single bead. Wells occupied by more than one bead cannot be used for the analysis contemplated by the present invention.

Ultimately, a uniform population of template nucleic acids is placed in one or more of the multiple wells, each well being placed on and thus corresponding to at least one ISFET. As noted above, preferably the wells contain at least 10, at least 100, at least 1000, at least 10 ⁴ , at least 10 ⁵ , at least 10 ⁶ , or more copies of the same template nucleic acid. The same template nucleic acid means at least that the templates are identical with respect to the sequence. Most preferably, all template nucleic acids in one well are uniformly hybridized with the primer. Homogeneous hybridization of a template nucleic acid and a primer means that the primer hybridizes to the template at the same position as all other primer-template hybrids in the well (ie, in a sequence along the template complementary to the primer). It means to soy. Uniform placement of primers in all templates allows for co-ordinated synthesis of all new nucleic acid strands in the well, thereby resulting in a greater signal to noise ratio.

  Nucleotides are then added to the flow or sequentially to the flow chamber and thus to the wells by any other suitable method. Nucleotides can be added in any order if known to remain constant for simplicity throughout the run. Where the incorporation of nucleotides is based on detection of PPi and not the detection of pH changes due to PPi release, it is preferable to maintain a relatively constant level and concentration of nucleotides throughout the reaction and washing. One way to accomplish this is to add ATP to the wash buffer in such a way that dNTPs that flow into the well move ATP out of the well. ATP balances the ionic strength of dNTP entering the well and has a diffusion profile similar to dNTP. With this method, dNTP influx and efflux during sequencing reactions do not interfere with measurements in chemFET. The concentration of ATP used is on the order of the concentration of dNTP used.

  A typical sequencing cycle proceeds as follows: the flow chamber (and well) is washed with an ATP-containing wash buffer, and a first dNTP species (eg dATP) is introduced into the flow chamber (and well). , PPi release and detection (by any mechanism described herein), flow chamber (and well) washed with ATP containing wash buffer, flow chamber (and well) washed with apyrase containing wash buffer, flow Wash chamber (and well) with ATP-containing wash buffer and introduction of second dNTP species. This process continues until all four dNTPs (ie, dATP, dCTP, dGTP, and dTTP) have flowed through the chamber to be incorporated into the newly synthesized strand. This 4-nucleotide cycle may be repeated up to any number of times, including but not limited to 10, 25, 50, 100, 200 or more. The number of cycles is controlled by the length of the template to be sequenced and the need to replenish reaction reagents, particularly dNTP stock solution and wash buffer.

  As part of the sequencing reaction, dNTPs are sequenced in the 3 ′ end of the newly synthesized strand (or in the case of the first incorporated dNT, if their complementary nucleotides are present at the same position in the template nucleic acid. (Or “incorporated” as used herein). Incorporation of introduced dNTPs (and simultaneous release of PPi) thus indicates identification of the corresponding nucleotide of the template nucleic acid. If no change in electric field is detected by the ISFET, it can be concluded that dNTPs were not incorporated and the complementary nucleotide was not present at that position in the template. When a change in the electric field is detected, the introduced dNTP is incorporated into the newly synthesized chain. There is a positive correlation between dNTP uptake and PPi release and ISFET response, so that further quantification of the number of dNTPs taken up is possible. In other words, the voltage change recorded in the ISFET is related to the number of dNTPs captured. The result is that sequence information was not lost through sequencing of homopolymer stretches in the template (eg, poly A, poly T, poly C, or poly G). As an example, if the template nucleic acid contains the sequence 5 ′ CAAAAG 3 ′, the ISFET records a signal upon introduction of dCTP (eg, in millivolt changes), then records a larger signal upon introduction of dTTP, followed by Another signal is recorded when dGTP is introduced. The magnitude of the signal generated upon introduction of dCTP and dTTP is essentially equivalent and correlates with millivolt changes resulting from single nucleotide incorporation. The magnitude of the signal produced upon introduction of dTTP is larger than the signal from dNTP uptake. The magnitude of these signals is additive, and depending on the length of the homopolymer, the extension may not be readily apparent in the voltage versus time (or frame) plots (eg, FIGS. 71A-D, right hand side). There is sex. The signal can be measured using the peak intensity of the voltage versus time (or frame) plot or the area under the curve.

  Apyrase is an enzyme that breaks down residual non-incorporated nucleotides into monophosphate and releases inorganic phosphorus into the process. This is useful for degrading dNTPs that have not been taken up and / or in excess in any or all wells. It is important to remove excess and / or unreacted dNTPs from any and all wells by washing prior to the next dNTP introduction. Thus, adding an apyrase during the synthesis reaction and during the introduction of different dNTPs is useful to remove excess dNTPs that would otherwise obscure the sequence data.

  Additional sequencing reaction reagents, such as those described above, may be introduced throughout the reaction, although in some cases this is not necessary. For example, additional polymerase, DTT, SBB, etc. may be added as needed.

  Thus, the present invention contemplates performing multiple different sequencing reactions simultaneously. Multiple identical sequencing reactions occur simultaneously in each occupied well. This simultaneous and identical dNTP incorporation within each well increases the signal to noise ratio, thereby allowing detection of sequencing reaction byproducts. By performing the sequencing reaction simultaneously in a plurality of wells, a plurality of different sequencing reactions are performed simultaneously.

  The sequencing reaction can be performed at a certain temperature range. Typically, the reaction is conducted in the range of 30-60 ° C, 35-55 ° C, or 40-45 ° C. The reaction is preferably carried out at a temperature that prevents the formation of secondary structures with the nucleic acid. However, this must be balanced between the binding of the primer (and the newly synthesized strand) to the template nucleic acid and the reduction of the apyrase half-life at elevated temperatures. A suitable temperature is about 41 ° C. Solutions containing wash buffer and dNTP solution are generally warmed to these temperatures to prevent the well temperature from changing. However, the apyrase-containing wash buffer is preferably maintained at a lower temperature to extend the half-life of the enzyme. Typically, the solution is maintained at about 4-15 ° C, and more preferably 4-10 ° C.

  Nucleotide incorporation reactions can occur very quickly. As a result, in some instances it may be desirable to slow down the reaction to ensure maximum data acquisition during the reaction. The diffusion of reagents and / or by-products can be slowed in many ways, including but not limited to adding packing beads to the wells. Packing beads also tend to increase reagent concentration and / or by-products on the chemFET surface, thereby increasing the potential for the signal. The presence of packing beads generally allows longer times for the sample (eg, 2-4 times).

  Data capture rates can vary, for example anywhere from 10 to 100 frames per second, and which rate is used is governed, at least in part, by the size of the wells and the presence of packing beads. Smaller well sizes generally require higher data capture rates.

In some aspects of the invention where the top surface of the well is open and communicates with fluid throughout the chip, the released PPi or other by-product (eg, H ⁺ ) is diffused out of the well. It is important to detect before going. Diffusion of any reaction by-product exiting the well leads to false negatives (because no by-products are detected in that well), leading to the possibility of false positives adjacent to or downstream of the well, and therefore Should be avoided. Packing beads can also help reduce the degree of diffusion and / or crosstalk between wells.

  Thus, in some embodiments, packing beads are used in addition to nucleic acid loaded beads. The packing beads may be magnetic (including superparamagnetism), but are not limited thereto. In some embodiments, the packing beads and capture beads are made of the same material (eg, both magnetic, both are polystyrene, etc.), while in other embodiments they are different materials (eg, the packing beads are polystyrene). Yes, the capture beads are magnetic). Packing beads are generally smaller than capture beads. The dimensional difference may vary and may be 5 times, 10 times, 15 times, 20 times or more. As an example, 0.35 μm diameter packing beads were used with 5.91 μm capture beads. Such packing beads are commercially available from sources such as Bang Labs. The arrangement of the packing beads relative to the capture beads may be different. For example, the packing beads surround the capture beads, thereby preventing the capture beads from contacting the ISFET surface. As another example, packing beads are loaded into the wells following the capture beads, where the capture beads are in contact with the ISFET surface. The presence of packing beads between the capture beads and the ISFET surface slows the diffusion rate of sequencing byproducts such as PPi, thereby facilitating data acquisition.

  The present invention further uses packing bead or chemFET surface modification (as described herein) to prevent the chemFET surface from contacting and interfering with the template nucleic acid bound to the capture bead. Intended. A layer of packing beads with a depth or height of 0.1-0.5 μm excludes this interaction.

  The sequencing reaction proceeds by analyzing the array and determining the position of the beads. In the absence of flow, the background signal (ie noise) is below 0.25 mV, but in the presence of DNA-loaded capture beads, the signal is found to increase to about 1.0 mV = / − 0.5 mV. It was done. This increase is sufficient to allow determination of wells with beads.

  The present invention further contemplates kits that include various reagents necessary to perform a sequencing reaction and instructions for use according to the methods described herein.

  One preferred kit includes one or more containers containing wash buffers and one or more containers each containing one of the following reagents: dATP buffer, dCTP buffer, dGTP buffer. Solution or dTTP buffer, dATP, dCTP, dGTP and dTTP stock solution, apyrase, SSB, polymerase, packing beads and optionally pyrophosphatase. Importantly, the kit contains only natural dNTPs.

  It will be appreciated that interactions between receptors and ligands or between two members of a binding pair or between elements of a molecular complex can also be detected using a chemFET array. Examples of such interactions include hybridization between nucleic acids, protein-nucleic acid binding, protein-protein binding, enzyme-substrate binding, enzyme-inhibitor binding, antigen-antibody binding, and the like. Any binding or hybridization event that causes a change in semiconductor charge density at the FET interface and thus changes the current flowing from the power source to the drain of the sensor described herein is detectable in accordance with the present invention.

  In these embodiments, the passivation layer (or potentially also the intermediate layer covered by the passivation layer) is a nucleic acid (eg, DNA, RNA, miRNA, cDNA, etc.), antigen (of any nature). Good), proteins (eg, enzymes, cofactors, antibodies, antibody fragments, etc.) and the like. The attachment of these entities to the passivation layer may be direct or indirect (eg, using a bifunctional linker that binds to both the passivation layer reactive group and the entity to be attached).

  As an example, a reactive group such as an amine or thiol group can be added to the nucleic acid at any nucleotide during synthesis to provide a point of attachment for the bifunctional linker. Other examples include conjugation-competent reagents such as Uni-Link AminoModifier, 3′-DMT-C6-Amine-ON CPG, AminoModifier II, N-TFA-C6-AminoModifier, C6-ThiolModifier, C6. -Nucleic acids may be synthesized by incorporating disulfide phosphoramidite, C6-disulfide CPG (Clontech, Palo Alto, CA) and the like. Other methods of attaching nucleic acids are described below.

  In one aspect of the invention, a chemFET array is provided in combination with a nucleic acid array. Nucleic acids in the form of short nucleic acids (eg, oligonucleotides) or long nucleic acids (eg, full length cDNA) can be provided on the chemFET surface of the arrays described herein. Nucleic acid arrays generally comprise a plurality of physically defined regions (eg, “spots”) on a flat surface, each having one, and more preferably two or more nucleic acids attached thereto. Nucleic acids are usually single stranded. The nucleic acids bound to a given spot are usually the same. In the context of oligonucleotide arrays, these nucleic acids may be less than 100 nucleotides long (including about 10, 20, 25, 30, 40, 50, 60, 70, 80, 90 or 100 nucleotides long). When the array is used to detect certain genes (including mutations in such genes or the level of expression of such genes), the array contains a number of oligonucleotides each containing oligonucleotides over a defined potentially different gene sequence. May include spots. These spots are then placed over a flat surface to eliminate position related effects in hybridization and array reading means.

  The array is contacted with the sample to be tested. The sample may be a genomic DNA sample, a cDNA sample from a cell, a tissue or mass (eg, a tumor), a population of cells grown on the array, which may be in a two-dimensional array, etc. corresponding to the underlying sensor array. It's okay. Such arrays are therefore mutations (eg, deletions, additions, substitutions, etc.) within a particular gene and include a single nucleotide polymorphism to determine the presence and / or level of that particular gene or its expression. , But not limited thereto).

  Binding or hybridization of sample nucleic acid and immobilized nucleic acid is generally performed under stringent hybridization conditions, a term understood in the art (see, eg, Sambrook et al. “Maniatis”). Examples of relevant conditions include (in order of increasing stringency): incubation temperatures 25 ° C., 37 ° C., 50 ° C. and 68 ° C., buffer concentrations 10 × SSC, 6 × SSC, 4 × SSC, 1 × SSC 0.1 × SSC (where SSC is 0.15 M NaCl and 15 mM citrate buffer) and their equivalents using other buffer systems, formamide concentrations 0%, 25%, 50% , And 75%, incubation time 5 minutes to 24 hours, 1, 2 or more wash steps, wash incubation time 1, 2, or 15 minutes, and wash solution 6 × SSC, 1 × SSC, 0.1 × SSC, Or deionized water. For example, hybridization is performed with 50% formamide and 4 × SSC, then washed with 2 × SSC / formamide at 50 ° C. and 1 × SSC.

Nucleic acid arrays include those in which nucleic acids, such as cDNAs already formed, are deposited (or “spotted”) at specific locations on the array. Nucleic acids are applied to the surface, for example, piezoelectric deposition, UV cross-linking of nucleic acids to polymer layers such as, but not limited to, poly-L-lysine or polypyrrole, as described in published US patent application 2003/0186262. Directly bonded to silicon-coated SiO ₂ and treated with a silanized chemFET surface (eg 3-aminopropyltriethoxysilane (APTES) as described in Uslu et al. Biosensors and Bioelectronics 2004, 19: 1723-1731). Can be spotted by direct binding to the surface.

  Nucleic acid arrays also include those in which nucleic acids (eg, oligonucleotides of known sequence) are synthesized directly into the array. Nucleic acids can be synthesized on arrays using, but not limited to, art-recognized techniques such as: printing with fine point pins on glass slides, pre-made masks Using photolithography, photolithography using dynamic micromirror devices (eg DLP mirrors), inkjet printing, or electrochemical methods on microelectrode arrays. See also Nuwaysir et al. 2002 "Gene expression analysis using oligonucleotide arrays produced by maskless photolithography.". Genome Res 12: 1749-1755. Commercial sources of this latter type of array include Agilent, Affymetrix and NimbleGen.

  Thus, the chemFET passivation layer may be coated with an intermediate layer of reactive molecules (and thus reactive groups) to which the nucleic acid binds and / or from which the nucleic acid is synthesized.

  The present invention contemplates combining such nucleic acids with chemFET arrays, particularly with the “large scale” chemFET arrays described herein. ChemFET / nucleic acid arrays can be used for a variety of applications, some of which do not require wells (or microwells or reaction chambers as used interchangeably herein). One or more flow chambers are placed on the array for analysis to be performed in the flow, including “closed” systems (ie, where the flow of reagents and wash fluids is automated). Touching. The use of multiple flow chambers allows multiple samples (or nucleic acid libraries) to be analyzed simultaneously. There may be 2, 3, 4, 5, 6, 7, 8, 9, 10 or more flow chambers. This configuration applies to other biological arrays as well, including those described herein, such as protein arrays, antibody arrays, enzyme arrays, chemical arrays, and the like.

  Because binding events between binding partners or elements of the complex are detected electronically via the underlying FET, such assays can be performed without the need to manipulate (eg, externally label) the sample to be tested. Can be implemented. This is advantageous because such manipulation always results in sample loss and generally increased time and manipulation. In addition, the method can test the binding reaction in real time.

  Protein arrays for use in combination with the chemFET of the present invention are also contemplated. A protein array includes proteins or peptides or other amino acids that contain biological moieties bound to a flat surface in a systematic and predetermined manner. Such proteins include, but are not limited to, enzymes, antibodies and antibody fragments or antibody mimics (eg, single chain antibodies).

In one embodiment, the protein array may include a plurality of different proteins (or other amino acids that include biological moieties). Each protein, and preferably a plurality of proteins, is present in a predetermined region or “cell” of the array. This region (or cell) is aligned with the sensors in the sensor array such that there is one sensor for each region (or cell). Multiple proteins within a single region (or cell) may vary depending on the size of the protein and the size of the region (or cell) and are at least 10, 50, 100, 500, 10 ³ , 10 ⁴ , Or more than that, but not limited thereto. Array itself may comprise any number of cells, including but contains at least 10, 10 ^{2, 10 3,} 10 ^4, 10 ^5, 10 6, 10 ^7, or more, but limited to Not. In one application, the array is exposed to a sample that contains or is suspected of containing an analyte that binds to the protein. The analyte may be a substrate or inhibitor if the protein is an enzyme. Analytes can be any molecule that binds to a protein, including other proteins, nucleic acids, chemical species (synthetic or natural), and the like.

  As with the nucleic acid arrays contemplated herein, reading from a protein array is a change in current through the chemFET, and therefore, in these array methods, an additional step of labeling and / or label detection. Is understood to be unnecessary.

  In other embodiments, the protein array may include multiple identical proteins (or other amino acids that contain biological moieties). The same protein may be uniformly distributed on a flat surface, or they may be organized into discrete areas (or cells) of this surface. In these latter aspects, the regions (or cells) are aligned with the sensors of the sensor array in such a way that there is one sensor in each region (or cell).

  The protein may be synthesized off-chip and then purified and attached to the array. Alternatively, the protein may be synthesized on-chip, similar to the nucleic acids described above. Protein synthesis using cell-free DNA expression or chemical synthesis is suitable for on-chip synthesis. Using cell-free DNA expression, once synthesized protein is attached to a solid support. Alternatively, the protein may be chemically synthesized using solid phase peptide synthesis on a solid support. Selective deprotection is performed using lithographic methods or SPOT synthesis. See at least MacBeath and Schreiber, Science, 2000, 289: 1760-1763 or Jones et al. Nature, 2006, 439: 168-174. See also Fodor et al US Pat. No. 6919211.

Chemical compound microarrays combined with chemFET arrays can also be envisaged. Chemical compound microarrays are immobilized on solid surfaces by covalently immobilizing compounds (eg, organic compounds) by various binding techniques on solid surfaces (also called “small molecule microarrays” in the literature). Organic compounds by spotting or drying compounds (eg organic compounds) without solubilization (also referred to in the literature as “microarray compound screening (μARCS)”) or in homogeneous solution without immobilization and drying effects (Commercialized as DiscoveryDot ^™ technology by Reaction Biology Corporation).

  Tissue microarrays in combination with chemFET arrays are further contemplated by the present invention. Tissue microarrays are described in further detail in Battifora Lab Invest 1986, 55: 244-248; Battifora and Mehta Lab Invest 1990, 63: 722-724; and Kononen et al. Nat Med 1998, 4: 844-847. ing.

  The configurations of chemFET arrays and biological or chemical arrays are similar in each example, and a description of one combinatorial array can be found elsewhere in this specification or others known in the art. Will be applied to.

  In yet another aspect, the present invention is placed in contact with a cell culture (eg, a two-dimensional cell culture) (see, eg, Baumann et al. Sensors and Actuators B 55 1999 77:89), and a chemFET array. An analysis of the tissue section that was performed is also contemplated. As an example, a brain slice can be placed in contact with the chemFET array of the present invention, and for example, without limitation, changes in the slice in the presence or absence of a stimulus such as a neurotoxin can be detected. Neural processes and / or stimulation transduction can thus be analyzed. In these embodiments, the chemFET is manipulated by detecting calcium and / or potassium flux through the passivation layer itself or through receptors for these ions coated on the passivation layer. Good.

  In yet another aspect, the present invention contemplates in vivo use of a chemFET array functionalized as described herein or otherwise. Such an array is introduced into a subject (eg, in the brain or other area that receives ion flux) and then analyzed for changes based on the condition of the subject.

  Although several inventive aspects have been described and illustrated herein, one of ordinary skill in the art can readily perform the functions described herein and / or result and / or one or more of Various other means and / or structures for obtaining advantages are envisioned, and each such variation and / or modification is considered to be within the scope of the embodiments of the invention described herein. More generally, those skilled in the art will readily appreciate that all parameters, dimensions, materials, and structures described herein are illustrative and that the actual parameters, dimensions, materials, and / or structures are Will be dependent on the particular application or applications in which it is used. Those skilled in the art will recognize, and be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. Accordingly, the foregoing aspects are shown by way of example only and, within the scope of the appended claims and their equivalents, the aspects of the present invention may be practiced differently from what is specifically described and claimed. Is understood. Inventive aspects of the present disclosure are directed to each individual feature, system, article, material, kit, and / or method described herein. Further, any combination of two or more of such features, systems, articles, materials, kits, and / or methods is such that such features, systems, articles, materials, kits, and / or methods are mutually related. If not inconsistent, they are included within the scope of the invention of this disclosure.

  All definitions defined and used herein should be understood as governing the dictionary definition, the definition in the literature incorporated by reference, and / or the ordinary meaning of the defined term. is there.

  As used herein and in the claims, the indefinite articles “a” and “an” (ie singular nouns) are understood to mean “at least one” unless explicitly stated to the contrary. Should.

  As used herein and in the claims, the phrase “and / or” includes “of connected elements, ie, elements that are presented in a connected manner and in other cases in a disjoint manner. It should be understood to mean "one or both". Multiple elements listed with “and / or” should be construed in the same manner as “one or more” of the concatenated elements. Other elements than those specifically identified by the “and / or” clause may optionally be presented, whether related to or unrelated to the specifically identified element. Thus, as a non-limiting example, when referring to “A and / or B” when used in conjunction with an unconstrained word such as “includes”, in one aspect, only A (optionally B Other elements), in other embodiments, only B (optionally including elements other than A), and in yet other aspects, both A and B (optionally including other elements). be able to.

  As used herein in the specification and in the claims, “or” should be understood to have the same meaning as “and / or” as defined above. For example, when separating items in a list, “or” or “and / or” is inclusive. That is, it should be construed that it includes at least one of a number of elements or element lists, but may include more than one, and optionally include additional items not in the list. Terms that clearly indicate the opposite, such as “only one” or “exactly one”, or “consisting of” when used in the claims, refer to an exact element within a number of elements or element lists. Refers to containing one element. In general, as used herein, the term “or” follows an exclusive term such as “either”, “one”, “only one”, or “exactly one”. Should be interpreted as indicating an exclusive alternative (ie, “one or the other but not both”). “Consisting essentially of”, when used in the claims, should have its ordinary meaning as used in the field of patent law.

  As used herein and in the claims, the phrase “at least one” referring to a list of one or more elements is at least one element selected from any one or more elements in the list of elements Is meant to mean, but does not necessarily include at least one of every element specifically listed in the element list, and excludes any combination of elements in the element list That's not true. This definition also refers to whether an element other than the element specifically identified in the element list to which the “at least one” phrase refers relates to or is associated with the specifically identified element. Even without it, it is possible to exist at will. Thus, as a non-limiting example, “at least one of A and B” (or equivalently, “at least one of A or B”, or equivalently, “at least one of A and / or B”) In one embodiment, at least one A is said to optionally include more than one A, where B is absent (and optionally includes elements other than B), in other embodiments, at least one B Optionally including more than one B, where A is not present (and optionally includes elements other than A), and in yet other embodiments, at least one A, optionally more than one A. Including, and at least one B, optionally including more than one B (and optionally including other elements), and the like.

  Unless expressly stated to the contrary, in any method claimed herein that includes more than one step or action, the order of the method steps or actions is not necessarily to describe the method steps or actions. It should be understood that the order is not limited.

  In the claims and in the above specification, “include”, “include”, “carry”, “have”, “include”, “accompany”, “hold”, “consist of”, etc. All transitional phrases are to be understood as unconstrained, which means including but not limited to these. Only the transitional phrases “consisting of” and “consisting essentially of” are limited or semi-closed, respectively, as described in Section 21111.03 of the US Patent Office Patent Examination Manual. It is a transitional phrase.

The following is a proof-of-principle example of a principle for rapid sequencing of single stranded oligonucleotides using an ISFET array.

1.1 Binding of single-stranded oligonucleotides to streptavidin-coated magnetic beads Single-stranded DNA oligonucleotide templates with 5 'dual biotin tags (HPLC purification) and 20 base universal primers are available from IDT (Integrated DNA Technologies, Coralville, IN ).

  The template was 60 bases long and was designed to contain 20 bases complementary to the 20 base primer at the 3 'end (Table 1, italic). Lyophilized and biotinylated templates and primers were resuspended in TE buffer (10 mM Tris HCl, 1 mM EDTA, pH 8), 40 μM stock solution and 400 μM stock solution, respectively, and stored at −20 ° C. until use. .

For each template, 5.91 μm streptavidin-coated magnetic beads (Bangs Laboratories, Inc. Fishers, IN), stored at 4 ° C. as an aqueous buffer suspension (8.57 × 10 ⁴ beads / μl). 60 μl was prepared by washing 3 times with 120 μl bead wash buffer, then templates 1, 2, 3 and 4 (T1, T2, T3, T4: Table 1) with biotin at the 5 ′ end were prepared. And incubated.

Due to the strong covalent affinity of streptavidin for biotin (Kd: -10-15), these magnetic beads are used to immobilize the template on a solid support, as described below. The reported binding capacity of these beads for free biotin is 0.650 pmol / 1 μl bead stock solution. For small (<100 bases) biotinylated ssDNA templates, conservatively calculated, 9.1 × 10 ⁵ templates can be bound per bead. The beads are easily concentrated using a simple magnet, similar to using a Dynal Magnetic Particle Concentrator or MPC-s (Invitrogen, Carlsbad, CA). MPC-s was used in the described experiment.

  Using MPC-s, the beads were concentrated for 1 minute between each wash, and then buffer was added to resuspend the beads. After the third wash, the beads were resuspended in 120 μl of bead wash buffer plus 1 μl of each template (40 μM). The beads were incubated for 30 minutes with rotation (Labquake Tube Rotator, Barnstead, Dubuque, IA). After incubation, the beads were then washed 3 times with 120 μl annealing buffer (20 mM Tris HCl, 5 mM magnesium acetate, pH 7.5) and resuspended in 60 μl of the same buffer.

1.2. Sequencing Primer Immobilization The immobilized template bound to the 5.91 μm magnetic beads at the 5 ′ end was then annealed to a 20 base primer complementary to the 3 ′ end of the template (Table 1). Add 1.0 μl aliquot of 400 μM primer stock solution, 20-fold excess of primer to immobilized template, then incubate beads and template with primer for 15 minutes at 95 ° C., then The temperature was slowly lowered to room temperature. The beads were then used with MPC-s as described above to 120 μl of 25 mM Tricine buffer (25 mM Tricine, 0.4 mg / ml PVP, 0.1% Tween 20, 8.8 mM Magnesium acetate, pH 7 8) and washed 3 times. The beads were resuspended in 25 mM Tricine buffer.

1.3. Incubation of hybridized template / primer with DNA polymerase The template / primer hybrid can be converted to polymerase as described essentially in Margulies et al. Nature 2005 437 (15): 376-380 and accompanying supplemental material. Incubated with.

2. Loading the Prepared Test Sample onto the ISFET Sensor Array The dimensions and density of the ISFET array and the microfluidics thereon may vary depending on the application. A non-limiting example is a 512 × 512 array. Each grid of this array (which will be 262144) has a single ISFET. Each grid also has a well (or is used herein as synonymous with “microwell”) located thereon.

  The well (or microwell) may have any shape including cylindrical, conical, square, rectangular and the like. In one exemplary form, the well is a 7 × 7 × 10 μm square well. The distance from the center of the well to the center is referred to herein as “pitch”. The pitch can be any distance, but a short pitch is preferred to accommodate as large an array as possible.

  The pitch may be less than 50 μm, less than 40 μm, less than 30 μm, less than 20 μm or less than 10 μm. In one embodiment, the pitch is about 9 μm. The entire chamber above the array (in which the wells are placed) may have a volume of about 30 μl or less, about 20 μl or less, about 15 μl or less, or about 10 μl or less. These volumes therefore also correspond to the volume of solution in the chamber.

2.1 Loading Beads on “Open” System Beads with templates 1-4 were loaded onto the chip (10 μl of each template). Briefly, an aliquot of each template was added onto the chip using an Eppendorf pipette. The beads were then drawn into the well using a magnet.

2.2 Loading the beads in a “closed” system Load both the capture and packing beads with the flow. The accuracy in microliters relative to the volume of the bead solution and the placement of the bead solution through the fluid connection is achieved using a bead loading fitting, as shown in FIG. The bead load attachment includes a large reservoir (about 1 ml volume), a small reservoir (about 10 μl volume), and a microfluidic channel for manipulating a small volume of bead solution. This method also takes advantage of the microliter accuracy of fluid application that is possible with precision pipettes.

  The chip containing the ISFET array and flow cell is placed in the ZIF (zero insertion force) socket of the load fixture, and then the stainless steel capillary tube is connected to one port of the flow cell and the flexible nylon tubing of the other port. Install. Both materials are microfluidic channels (eg with an inner diameter of <0.01 ”).

  The bead load attachment consists of large and small reservoirs that attach to the end of the capillary. Fill a common plastic syringe with buffer and then connect to the free end of the nylon tubing. The lead wire protruding from the bottom of the chip is inserted into the upper socket of the fixture unit (not shown).

  As shown in FIG. 63, a syringe is pushed to inject a buffer solution, and the buffer solution passes through the pipe, crosses the flow cell (and the chip surface), and goes up through the capillary tube. This process is called priming and ensures that no air enters the fluid path. Buffer is infused until the liquid level is visible through the clear large reservoir to the top of the small reservoir.

  Next, as shown in FIG. 64, the solution containing the nucleic acid-coated beads is applied to a small reservoir using a precision pipette. This application produces large droplets on the reservoir. The volume of solution added is equal to the volume of the flow chamber (eg, on the order of 10 μl), and the concentration of this bead is the desired transfer rate to the flow cell when initially added to the buffer dose present in the small reservoir. A concentration that produces a bead concentration.

  Carefully withdraw the syringe until the pipette is in place and the droplet has retracted over the small reservoir and is again visible through the clear large reservoir, as shown in FIG. Because the microfluidic channel extending down from the small reservoir is very small (eg, about 0.01 "in diameter), there is little mixing between the bead solution and the buffer in the flow path during this process.

  At this point, the bead solution is loaded into the flow path but has not yet reached the location of the flow cell. Before transferring the volume of bead solution in the bead solution plug or flow path, the solution in the small reservoir is washed. Initially, as shown in FIG. 66, approximately 1 ml of buffer is injected into the large reservoir, effectively diluting the bead solution left in the small reservoir. The solution is then pipetted with the pipette tip positioned along the bottom edge of the large reservoir. The level of the solution in the small reservoir is kept at its maximum level as shown in FIG.

  A constant volume of buffer is then added as droplets onto a small reservoir, similar to the previous application of bead solution as shown in FIG. The volume of this solution is equal to the volume of the flow path between the small reservoir and the flow chamber of the flow cell (ie microfluidic channel volume + capillary volume + flow cell volume in front of the flow chamber). Again, the syringe is withdrawn until the droplet retracts over the small reservoir as shown in FIG. Here, the bead solution plug is loaded into the flow chamber of the flow cell.

  Here, the load fixing tool is lifted and placed on a pyramid-shaped base including a magnet at the apex thereof as shown in FIG. The magnet pulls the beads from the bead solution to the microwells of the chip. After a few seconds, remove the fixture from the base. The entire process except initial priming with fluid buffer can be repeated to load small packing beads into the microwells as needed.

  It will be appreciated that there may be other ways to draw the beads into the wells of the flow chamber, such as centrifugation or gravity. The present invention is not limited in this respect.

3. DNA Sequencing with ISFET Sensor Array 3.1 DNA Sequencing in an “Open” System The results presented are representative of experiments performed in an “open” system (ie, this experiment is an ISFET chip) Are placed on the platform of the ISFET device, then each nucleotide (5 μl, resulting 6.5 μM each) is manually transferred in the following order: dATP, dCTP, dGTP and dTTP (100 mM stock solution, Pierce, Milwaukee, WI) The given nucleotide was pipetted into the liquid already on the chip surface and data was collected from the chip at a rate of 2.5 mHz). This resulted in data collection of approximately 18 frames / second in 7.5 seconds. The data was then analyzed using Lab View.

  For certain template sequences, the addition of dATP resulted in a 4 base extension of template 4. The addition of dCTP resulted in a 4 base extension of template 1. The addition of dGTP resulted in the extensions shown in Table 2 for templates 1, 2 and 4, and the addition of dTTP resulted in a runoff (as shown, extension of the whole template).

  FIG. 71 (AD) shows an extension reaction. In the left panel, all pixels for one snapshot in time are shown, the right side shows a graph of mV vs. time for the four selected pixels from the left set. White arrows indicate active pixels that are undergoing stretching. In run-off (FIG. 71D), in addition to the marked wells of FIG. 71C, the additional arrow indicates that no elongation occurred after the addition of dGTP, but rather a well where elongation occurred after the addition of dATP, Later, it was seen again during the runoff.

  If the method is performed in a non-automated manner (ie, without automatic flow and reagent introduction), each well preferably contains an apyrase to degrade unincorporated dNTPs. It is to be understood that the apyrase can be replaced with other compounds (or enzymes) capable of degrading dNTPs in this aspect or in any other aspect described herein.

3.2 Sequencing with microfluidics on sensor chips Sequencing in a flow regime is an extension of the open application to the incorporation of nucleotide reagents into DNA. Rather than adding the reagent into the bulk solution of the ISFET chip, the reagent is flowed across the chip surface in a continuous fashion, extending a single DNA base (s) at a time. dNTP flows from dTTP first and then continuously with dATP, dCTP and dGTP. Due to the laminar nature of the fluid motion on the chip, diffusion of nucleotides into the microwells and ultimately around the nucleic acid loaded beads is the primary mechanism of delivery.

The flow regime also ensures that most of the nucleotide solution is washed away during application. This involves rinsing the chip with each buffer stream with buffer and apyrase solution. Nucleotides and wash solutions are stored in the system's chemical bottles and flowed over the chip using fluid piping and automatic valve systems.
The ISFET chip is activated to sense chemical products of DNA extension during the nucleotide stream.

Claims (16)

A method for sequencing a nucleic acid, comprising:
Placing a plurality of templates nucleic acid in each reaction chamber of the plurality of reaction chambers, wherein the plurality of reaction Chang Ba, at least one chemical-sensitive field effect transistor having a floating gate for each reaction chamber ( chemFET ) ,
Sequentially one or more known nucleotide triphosphates by incorporating the end of a sequencing primer which hybridizes with at least one template nucleic acid, at least one template nucleic acid disposed on at least one reaction chamber Synthesizing a new nucleic acid strand from where a new nucleic acid strand is synthesized at least in part in the presence of a polymerase; and
Detecting the incorporation of one or more known nucleotide triphosphates by a change in current in at least one chemFET having a floating gate;
Only including, a floating gate, is coupled to the reaction chamber through a passivation layer which is sensitive to hydrogen ions, said method.   2. The method of claim 1, wherein the change in current is indicative of uptake of at least one nucleotide triphosphate.   2. The method of claim 1, wherein the change in current is indicative of incorporation of multiple identical nucleotide triphosphates.   The method of claim 1, wherein a plurality of template nucleic acids are bound to the beads and at least one bead is disposed in each of the plurality of reaction chambers.   The method of claim 1, wherein the synthesis step is performed from the 3 ′ end of the sequencing primer. The method of claim 1, wherein the plurality of reaction chambers comprises at least 10 ⁵ reaction chambers.   The method of claim 1, further comprising synthesizing a plurality of new nucleic acid strands from a corresponding plurality of template nucleic acids disposed in at least one reaction chamber.   The method of claim 1, wherein the change in current is due to a change in threshold voltage of at least one chemFET resulting from incorporation of one or more known nucleotide triphosphates. A method for sequencing a nucleic acid, comprising:
Placing a plurality of templates nucleic acid in each reaction chamber of the plurality of reaction chambers, wherein the plurality of reaction Chang Ba, at least one chemical-sensitive field effect transistor having a floating gate for each reaction chamber ( chemFET ) ,
Sequentially one or more known nucleotide triphosphates by incorporating the end of a sequencing primer which hybridizes with at least one template nucleic acid, at least one template nucleic acid disposed on at least one reaction chamber Synthesizing a new nucleic acid strand from where a new nucleic acid strand is synthesized at least in part in the presence of a polymerase; and
Detecting the incorporation of one or more known nucleotide triphosphates by a change in current in at least one chemFET having a floating gate, wherein the center-to-center distance between adjacent reaction chambers is about 1 μm to about 10 μm Selected from a range of
Only including, a floating gate, is coupled to the reaction chamber through a passivation layer which is sensitive to hydrogen ions, said method. The method of claim 9 , wherein the plurality of reaction chambers comprises at least 10 ⁵ reaction chambers. 10. The method of claim 9 , wherein the change in current is due to a change in threshold voltage of at least one chemFET resulting from incorporation of one or more known nucleotide triphosphates. A method for sequencing a nucleic acid, comprising:
Placing a plurality of templates nucleic acid in each reaction chamber of the plurality of reaction chambers, wherein the plurality of reaction Chang Ba, at least one chemical-sensitive field effect transistor having a floating gate for each reaction chamber ( chemFET ) ,
Sequentially one or more known nucleotide triphosphates by incorporating the end of a sequencing primer which hybridizes with at least one template nucleic acid, at least one template nucleic acid disposed on at least one reaction chamber Synthesizing a new nucleic acid strand from where a new nucleic acid strand is synthesized at least in part in the presence of a polymerase; and
Detecting changes in the level of sequencing by-products as an indicator of the incorporation of one or more known nucleotide triphosphates;
Only including, a floating gate, is coupled to the reaction chamber through a passivation layer which is sensitive to hydrogen ions, said method. 13. The method of claim 12 , wherein the sequencing byproduct is inorganic pyrophosphate (PPi). 14. The method of claim 13 , wherein the PPi binds to the PPi receptor in at least one reaction chamber. The method of claim 12 , wherein the plurality of reaction chambers comprises at least 10 ⁵ reaction chambers. The method of claim 12 , wherein detecting a change in the level of a sequencing byproduct comprises detecting a threshold voltage of at least one chemFET.

Images